# AETRE & The Governed Agent (oci · ghcr.io/grayclayton/aetre-mcp:0.8.1)

Decision-theoretic triage and fail-closed agent gating: 32 tools over stdio.

- Trust score: 36/100 (low)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-21

## Components

- oci · `ghcr.io/grayclayton/aetre-mcp:0.8.1`: 36/100 (this document), [markdown](https://verifymcp.io/servers/grayclayton-aetre-mcp/ghcr-io-grayclayton-aetre-mcp-0-8-1.md), [page](https://verifymcp.io/servers/grayclayton-aetre-mcp/ghcr-io-grayclayton-aetre-mcp-0-8-1)

## Channel facts

- Registry: `oci`
- Package: `ghcr.io/grayclayton/aetre-mcp:0.8.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-09-21.

- **Supply Chain Security**: 0/100
  - Malware scan not yet available for this package.
  - Known CVEs could not be checked: this artifact ships no SBOM, so there is no dependency list to read. Publishing one would let us assess it.
  - Install-script risk not yet assessed.
  - Dependency health could not be checked: this artifact ships no SBOM, so there is no dependency list to read. Publishing one would let us assess it.
- **Provenance & Transparency**: 35/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - License check failed: no license is declared.
  - Actively maintained (last published 0 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 78/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 5413 tokens (~146/item across 37 items; 33 tools + 4 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 33 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 34 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 20/100
  - Spec-recency check failed: implements MCP spec 2024-11-05; the latest is 2026-07-28.

**Unverified: 2 categories.** Categories scored 0 because we could not verify them: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the AETRE & The Governed Agent MCP server?

AETRE & The Governed Agent runs locally as a container image, launched with docker run --rm -i ghcr.io/grayclayton/aetre-mcp:0.8.1. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add grayclayton-aetre-mcp -- docker run --rm -i ghcr.io/grayclayton/aetre-mcp:0.8.1
```

### Cursor

```json
{
  "mcpServers": {
    "grayclayton-aetre-mcp": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/grayclayton/aetre-mcp:0.8.1"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "grayclayton-aetre-mcp": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/grayclayton/aetre-mcp:0.8.1"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add grayclayton-aetre-mcp -- docker run --rm -i ghcr.io/grayclayton/aetre-mcp:0.8.1
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "grayclayton-aetre-mcp": {
      "type": "local",
      "command": [
        "docker",
        "run",
        "--rm",
        "-i",
        "ghcr.io/grayclayton/aetre-mcp:0.8.1"
      ],
      "enabled": true
    }
  }
}
```

### Hermes

```yaml
mcp_servers:
  grayclayton-aetre-mcp:
    command: "docker"
    args: ["run", "--rm", "-i", "ghcr.io/grayclayton/aetre-mcp:0.8.1"]
```

### Netclaw

```json
{
  "McpServers": {
    "grayclayton-aetre-mcp": {
      "Transport": "stdio",
      "Command": "docker",
      "Arguments": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/grayclayton/aetre-mcp:0.8.1"
      ]
    }
  }
}
```

### Other

```json
{
  "mcpServers": {
    "grayclayton-aetre-mcp": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "ghcr.io/grayclayton/aetre-mcp:0.8.1"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-09-20 (score 36)

First indexed and scored.

## MCP tools (33)

### `aetre_system_catalog` (~76 tokens)

Comprehensive system introspection returning AETRE architecture, bundled synthetic fixtures, optional data adapters, connectors, mathematical tools, and institutional tiers.

Input parameters:

- `api_key` (string): Optional AETRE API or license key for tier verification.
- `query_type` (string): Category of system capability metadata to inspect. Defaults to 'all'.

### `aetre_triage_proposal` (~126 tokens)

Applies transparent, uncalibrated lexical routing indicators to proposal text, then calculates a VOI index and demonstration stage route (FAST-PASS, FAST-REJECT, or DEEP REVIEW). Not a validated estimate of scientific quality.

Input parameters:

- `api_key` (string): Optional AETRE API or license key.
- `selection_boundary` (number): Decision cutoff boundary for acceptance. Defaults to 1.2.
- `text` (string, required): The abstract, executive summary, or proposal body text to evaluate.
- `title` (string): Optional title of the proposal.

### `aetre_calculate_voi` (~196 tokens)

Calculates the exact Bayesian Value of Information (VOI) for crossing a top-K selection boundary under Gaussian conjugate updates.

Input parameters:

- `api_key` (string): Optional AETRE API or license key.
- `posterior_mean` (number, required): Current expected latent quality (mu).
- `posterior_variance` (number, required): Current epistemic uncertainty / variance (sigma^2).
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `review_cost` (number): Cost of conducting the review. Defaults to 0.5.
- `selection_boundary` (number, required): The threshold quality cutoff for acceptance (tau).
- `signal_noise` (number): Standard deviation of the additional review signal. Defaults to 0.8.

### `aetre_check_governor` (~163 tokens)

Evaluates evaluator queue load using Kingman's Heavy-Traffic approximation and returns governor throttle recommendations when utilization exceeds rho >= 0.85.

Input parameters:

- `api_key` (string): Enterprise license key required.
- `arrival_rate` (number, required): Arrival rate of submissions (lambda), items per period.
- `cv_arrivals` (number): Coefficient of variation of arrivals (c_a). Defaults to 1.0.
- `cv_service` (number): Coefficient of variation of review duration (c_s). Defaults to 1.0.
- `service_rate` (number, required): Review capacity of the committee/system (mu), items per period.
- `target_utilization` (number): Target sustainable utilization ceiling (rho_target). Defaults to 0.85.

### `aetre_exploration_audit` (~123 tokens)

Calculates the unbiased Horvitz-Thompson exploration audit estimator (H_hat_D) and 95% confidence intervals on deprioritized candidates to catch false negative breakthroughs.

Input parameters:

- `api_key` (string): Enterprise license key required.
- `audited_high_value_found` (integer, required): Number of high-value unconventional breakthroughs found in the audit sample.
- `audited_sample_size` (integer, required): Number of randomly sampled candidates audited (m_D).
- `deprioritized_pool_size` (integer, required): Total size of the rejected or deprioritized candidate pool (N_D).

### `aetre_evaluate_staking` (~166 tokens)

Simulates submitter entry equilibrium under generative AI generation costs and refundable submission deposits to curb spam floods.

Input parameters:

- `acceptance_capacity` (integer, required): Total available acceptance slots (K), e.g. 200.
- `api_key` (string): Enterprise license key required.
- `generation_cost` (number, required): AI generation cost per candidate (c_gen), e.g. $0.05.
- `private_acceptance_value` (number, required): Submitter's private value of winning acceptance (V), e.g. $100.00.
- `submission_fee` (number, required): Required deposit or submission stake (c_sub), e.g. $5.00.
- `total_potential_applicants` (integer, required): Total potential applicant pool (N), e.g. 5000.

### `aetre_proposition_1_bound` (~169 tokens)

Calculates Proposition 1 theoretical recall ceiling R_N <= min(1, K_N / H_N) to determine if a pipeline is mathematically capacity-constrained.

Input parameters:

- `api_key` (string): Optional AETRE API or license key.
- `high_value_rate` (number, required): Prior fraction of high-value ideas in population (p_H), e.g. 0.067.
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `selection_capacity` (integer, required): Available selection capacity (K).
- `total_candidates` (integer, required): Total candidate arrival volume (N).

### `aetre_correlated_posterior_update` (~190 tokens)

Calculates Bayesian posterior mean and uncertainty under correlated multi-agent evaluator noise (rho_corr), preventing artificial overconfidence from redundant LLM outputs.

Input parameters:

- `api_key` (string): Enterprise license key required.
- `evaluations` (array, required): List of evaluator agent scores and noise standard deviations.
- `inter_agent_correlation` (number): Pairwise correlation coefficient between evaluator errors (rho in [0, 1)). Defaults to 0.5.
- `prior_mean` (number, required): Prior mean of candidate quality (mu_0).
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `prior_variance` (number, required): Prior variance of candidate quality (sigma_0^2).

### `aetre_heavy_tailed_voi` (~237 tokens)

Calculates Generalized Pareto / Heavy-Tailed Value of Information (VOI) to optimize selection pipelines for positive black swan breakthrough discovery.

Input parameters:

- `api_key` (string): Enterprise license key required.
- `posterior_mean` (number, required): Current expected candidate quality (mu).
- `posterior_variance` (number, required): Current epistemic uncertainty (sigma^2).
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `review_cost` (number): Cost of conducting review. Defaults to 0.5.
- `selection_boundary` (number, required): Threshold cutoff boundary for selection (tau).
- `signal_noise` (number): Noise standard deviation of additional deep review. Defaults to 0.8.
- `tail_index_alpha` (number): Pareto tail index alpha > 1.0 (e.g. 1.5 for heavy-tailed scientific/biotech innovation). Defaults to 1.5.

### `aetre_quadratic_staking` (~191 tokens)

Calculates super-linear anti-sybil staking deposit requirements (Stake(m) = S_0 * m^gamma) to deter mass AI spam submissions while preserving human entry.

Input parameters:

- `api_key` (string): Enterprise license key required.
- `base_fee` (number, required): Base deposit for a single submission (S_0), e.g. $5.00.
- `escalation_exponent` (number): Escalation exponent gamma >= 1.0 (e.g. 2.0 for quadratic escalation). Defaults to 2.0.
- `generation_cost` (number): AI generation cost per submission (c_gen). Defaults to 0.05.
- `private_acceptance_value` (number): Private monetary or prestige payoff if accepted (V). Defaults to 100.0.
- `submission_count` (integer, required): Total submissions attempted by the entity within the time window (m).

### `aetre_heterogeneous_queues` (~63 tokens)

Evaluates a multi-specialist heterogeneous reviewer network, identifying bottleneck domains and generating capacity rebalancing actions.

Input parameters:

- `api_key` (string): Enterprise license key required.
- `pools` (array, required): List of domain queues with arrival and service parameters.

### `aetre_author_preflight_benchmark` (~105 tokens)

Comprehensive pre-submission diagnostic scorecard for authors and researchers, calculating crowd novelty percentile, reviewer disagreement risk, and prescriptive refinement actions.

Input parameters:

- `api_key` (string): Optional AETRE Pro or Enterprise license key for unlimited checks.
- `selection_boundary` (number): Funding or acceptance cutoff threshold (tau). Defaults to 1.2.
- `text` (string, required): Full proposal abstract or summary.
- `title` (string): Proposal or paper title.

### `aetre_simulate_benchmark` (~231 tokens)

Runs a paired-cohort Monte Carlo simulation across all 4 screening regimes, comparing Quality Throughput, FDR, Unconventional Recall, and Human Reviews with central 95% run-to-run outcome intervals (not confidence intervals for the mean).

Input parameters:

- `acceptance_capacity` (integer): Number of acceptance slots K (default: 200).
- `ai_arrival_multiplier` (number): Multiplier for synthetic/AI flood regime (default: 5.0).
- `api_key` (string): Enterprise license key required.
- `baseline_arrivals` (integer): Baseline arrival volume N (default: 1000).
- `evaluation_budget` (number): Total available evaluation budget (default: 1000.0).
- `randomized_audit_budget_share` (number): Share of budget allocated to randomized Horvitz-Thompson exploration audits (default: 0.05).
- `replications` (integer): Number of Monte Carlo simulation replicates (default: 50).
- `unconventional_share` (number): Prior share of unconventional/novel ideas (default: 0.10).

### `aetre_batch_triage` (~101 tokens)

Batch applies disclosed, uncalibrated lexical indicators to a cohort, computing heuristic ranks, VOI ranks, and demonstration stream allocation (Stream A Fast-Reject, Stream B Deep Review, Stream C Fast-Pass).

Input parameters:

- `api_key` (string): Optional license key.
- `proposals` (array, required): List of proposals with title and text/abstract.
- `selection_boundary` (number): Cutoff threshold boundary (default: 1.2).

### `aetre_recall_scaling_curve` (~218 tokens)

Calculates the Proposition 1 theoretical recall decay curve across arrival expansion scales (e.g. 1x, 2x, 5x, 10x, 20x, 50x) demonstrating capacity collapse points.

Input parameters:

- `api_key` (string): Optional license key.
- `baseline_arrivals` (integer): Baseline candidate arrivals N (default: 1000).
- `high_value_rate` (number): Prior high-value fraction in population (default: 0.067).
- `multipliers` (array): List of arrival multipliers to sweep across (default: [1, 2, 5, 10, 20, 50]).
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `selection_capacity` (integer): Available selection capacity K (default: 200).

### `aetre_fit_boundary` (~210 tokens)

Fits the decision boundary to a venue's own calibration data by sweeping thresholds and ranking candidates by boundary VOI at each. The boundary determines whether triage beats chance, and a default carried from another corpus generally does not.

Input parameters:

- `api_key` (string): Optional license key.
- `budget` (integer): Review budget K the boundary is optimised for. Defaults to 200.
- `dataset` (string, required): Path to a JSON array of candidates: {id, split, label, pre_triage_data:{preliminary_mean, preliminary_variance}}. Required: fit against your own data.
- `grid_max` (number): Highest boundary to try. Defaults to 10.0.
- `grid_min` (number): Lowest boundary to try. Defaults to 1.0.
- `grid_step` (number): Step between candidate boundaries. Defaults to 0.25.
- `split` (string): Split to fit on. Defaults to 'calib'. Fit and evaluate on different splits.

### `aetre_heldout_backtest` (~184 tokens)

Runs a multi-policy held-out review allocation backtest across 8 triage policies under fixed review budget K, evaluating true decision flips, precision, recall, and paired bootstrap intervals.

Input parameters:

- `api_key` (string): Optional license key.
- `boundary` (number): Acceptance threshold boundary theta (default: 6.0).
- `budget` (integer): Fixed review capacity budget K (default: 50).
- `dataset` (string): Path to a JSON array of candidates to evaluate: {id, split, label, pre_triage_data:{preliminary_mean, preliminary_variance}}. Omit to run against a bundled six-candidate sample, which illustrates the…
- `split` (string): Evaluation split ('test', 'dev', 'calib', 'replication', 'all') (default: 'test').

### `aetre_calibrate_scorer` (~118 tokens)

Fits Platt logistic scaling on continuous model scores and binary labels, returning slope, intercept, Expected Calibration Error (ECE), and Brier score.

Input parameters:

- `api_key` (string): Optional license key.
- `iterations` (integer): Calibration optimization iterations (default: 500).
- `labels` (array, required): Binary ground-truth labels (0 or 1).
- `learning_rate` (number): Optimization learning rate (default: 0.05).
- `scores` (array, required): Raw continuous candidate scores or VOI values.

### `aetre_multi_attribute_voi` (~131 tokens)

Computes multi-attribute Bayesian Value of Information across orthogonal proposal evaluation dimensions (Novelty, Rigor, Impact, Feasibility), outputting composite VOI and optimal dimension-specific review targets.

Input parameters:

- `api_key` (string): Optional license key.
- `composite_threshold` (number): Composite decision threshold cutoff (default: 6.0).
- `dimensions` (array, required): List of evaluation dimensions with name, prior_mean, prior_variance, weight, and review_noise_sd.
- `review_cost_per_dim` (number): Marginal review cost per dimension (default: 1.0).

### `aetre_congestion_matching` (~134 tokens)

Optimizes reviewer-to-proposal assignment by maximizing domain/keyword affinity while enforcing Kingman queue utilization constraints (rho <= 0.85) on individual reviewer workloads.

Input parameters:

- `api_key` (string): Enterprise license key.
- `proposals` (array, required): List of candidate proposals with id, title, domain, voi_index, required_reviews, and keywords.
- `reviewers` (array, required): List of reviewer profiles with id, name, domain, capacity, current_load, service_rate, arrival_rate, and expertise_tags.
- `target_utilization` (number): Maximum allowed reviewer utilization target (default: 0.85).

### `aetre_sequential_stopping_rule` (~256 tokens)

Calculates optimal dynamic Bayesian stopping boundaries for sequential reviews (Accept, Reject, or Solicit More Reviews) based on posterior decision confidence and boundary VOI.

Input parameters:

- `api_key` (string): Optional license key.
- `confidence_threshold` (number): Target confidence probability to stop early (default: 0.90).
- `next_review_cost` (number): Cost of soliciting an additional review (default: 1.0).
- `next_review_noise_sd` (number): Expected noise SD of a future review (default: 0.80).
- `prior_mean` (number, required): Baseline prior mean quality (e.g. 5.0).
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `prior_variance` (number, required): Baseline prior epistemic variance (e.g. 1.0).
- `reviews` (array, required): Ordered sequence of completed reviewer scores with noise_sd and cost.
- `threshold` (number, required): Decision acceptance threshold cutoff (e.g. 6.0).

### `governed_bellman_triage` (~219 tokens)

Pillar I Bellman Governor: evaluates Bayesian dynamic programming stopping policy over multi-stage pass lattices under asymmetric loss stakes (L/R), returning optimal action (CONTINUE, HALT_AND_COMMIT, HALT_AND_REJECT), expected utility, VOI, and critical threshold p*.

Input parameters:

- `api_key` (string): Optional license key.
- `consecutive_passes` (integer): Number of consecutive test passes observed (default: 0).
- `defect_leakage` (number): Defect leakage rate q (default: 0.5875).
- `loss` (number): Defective candidate loss penalty L (default: 0.10).
- `max_stages` (integer): Maximum verification stages horizon H (default: 4).
- `prior` (number): Prior belief in conforming status (default: 0.50).
- `reward` (number): Conforming candidate net reward R (default: 0.02).
- `stage` (integer): Current verification stage index (default: 0).

### `governed_review_boundary` (~242 tokens)

Section 4.1 Tripartite Review Boundary: evaluates whether an autonomous coding candidate should be AUTO-admitted, sent to human REVIEW, or ABSTAINED based on reviewer effort cost and knapsack capacity shadow price lambda_K.

Input parameters:

- `api_key` (string): Optional license key.
- `belief` (number, required): Current posterior belief probability in [0, 1].
- `loss` (number): Defective loss penalty L (default: 0.10).
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `review_accuracy` (number): Probability reviewer correctly verifies valid code (default: 1.0).
- `review_cost` (number): Direct cost of human reviewer examination (default: 0.002).
- `reward` (number): Conforming net reward R (default: 0.02).
- `shadow_price_lambda` (number): Knapsack queue capacity congestion shadow price lambda_K (default: 0.0).

### `governed_knapsack_admit` (~147 tokens)

Pillar V Knapsack Queue Controller: packs candidate pull requests into the review queue under capacity budget K using the c-mu rule (density rho_i = E[U_i] / k_i) and calculates the dual capacity shadow price lambda_K.

Input parameters:

- `api_key` (string): Optional license key.
- `candidates` (array, required): List of candidate submissions with candidate_id, posterior_belief, and optional reward, loss, review_cost.
- `capacity_k` (number, required): Total reviewer capacity budget K (e.g. 3.0 review slots or hours).
- `review_cost_k` (number): Default review effort cost per candidate (default: 1.0).

### `governed_gate_pr` (~93 tokens)

Road A Tiered Verification Gate: evaluates host pre-checks (Tier 0 syntactic AST parsing and structural validation) on candidate Python code or pull request patches to short-circuit broken submissions before expensive container CI escalation.

Input parameters:

- `api_key` (string): Optional license key.
- `candidate_id` (string): Identifier for the pull request or patch.
- `code` (string, required): Source code or patch string to screen.

### `governed_evaluate_action` (~147 tokens)

Micro-level loss evaluation comparing autonomous execution against deferral to human review, under an asymmetric loss matrix weighted by irreversibility.

Input parameters:

- `action_name` (string): Identifier for the candidate agent action.
- `api_key` (string): Optional license key.
- `consequence_distribution` (object, required): Distribution over the consequences of acting autonomously.
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `review_cost` (number): Cost of deferring the action to a human reviewer.

### `governed_stopping_policy` (~125 tokens)

Solves the finite-horizon optimal stopping problem by backward induction: at each step the agent either takes the terminal payoff or pays the continuation cost for one more step of evidence.

Input parameters:

- `api_key` (string): Optional license key.
- `cost_per_step` (number): Cost charged for each additional step of evidence gathering.
- `horizon_steps` (integer): Number of steps in the lattice. Defaults to the length of terminal_payoffs.
- `terminal_payoffs` (array, required): Payoff from stopping at each step. The last value is carried forward if shorter than the horizon.

### `governed_invariant_check` (~106 tokens)

Evaluates declarative safety invariants against a runtime state snapshot before a state change is committed, returning every violated rule rather than the first.

Input parameters:

- `api_key` (string): Optional license key.
- `invariant_rules` (array, required): Rules of the form '<field> <op> <number>' with op in <=, <, >=, >, ==, != , or 'exists <field>'.
- `state_snapshot` (object, required): Flat object of runtime state fields to test.

### `governed_shadow_price` (~185 tokens)

Computes the marginal shadow price of scarce reviewer time (lambda_K) from queue backlog and review capacity under a heavy-tailed utility distribution, and the factor by which it elevates the admission cutoff.

Input parameters:

- `api_key` (string): Optional license key.
- `prior_source` (string): Where the estimates in this call came from, e.g. "measured: 2026 cycle", "expert elicitation", "assumed". Echoed in the result so a reader can tell a measurement from a guess.
- `queue_backlog` (number, required): Number of candidates awaiting review.
- `reviewer_headcount` (number, required): Number of available reviewers.
- `reviews_per_reviewer` (number): Review slots per reviewer in the period. Defaults to 8.
- `tail_index_alpha` (number): Pareto tail index of the utility distribution. Defaults to 1.25.

### `governed_recall_scaling` (~130 tokens)

Fits the recall saturation curve Recall(K) = 1 - exp(-gamma K) to observed budget and recall pairs by least squares, returning the fitted gamma and the budget beyond which marginal recall stops paying for itself.

Input parameters:

- `api_key` (string): Optional license key.
- `historical_budget_K` (array, required): Observed review budgets.
- `marginal_recall_threshold` (number): Marginal recall per unit budget below which spending stops. Defaults to 0.001.
- `recall_points` (array, required): Recall achieved at each budget (0-1), positionally paired with historical_budget_K.

### `governed_runtime_audit` (~107 tokens)

Generates a tamper-evident SHA-256 decision receipt over an execution's identifier, payload and timestamp, so a gated agent action can be verified after the fact.

Input parameters:

- `api_key` (string): Optional license key.
- `decision_payload` (object): The decision record to bind into the receipt.
- `execution_id` (string, required): Identifier of the agent execution being recorded.
- `include_posterior_trace` (boolean): Echo the posterior trace from the payload into the receipt.

### `aetre_investment_benchmark` (~210 tokens)

Runs the venture dealflow triage benchmark: generates a synthetic heavy-tailed cohort of deals and compares status-quo preliminary-score screening against AETRE heavy-tailed VOI triage under a fixed diligence budget.

Input parameters:

- `api_key` (string): Optional license key.
- `diligence_budget` (integer): Deals that can be taken to full diligence. Defaults to 50.
- `hours_per_diligence` (number): Analyst hours consumed per deal diligenced. Defaults to 20.0.
- `n_deals` (integer): Deals in the synthetic cohort. Defaults to 1000, capped at 20000.
- `selection_boundary` (number): Preliminary score boundary for selection. Defaults to 6.0.
- `tail_alpha` (number): Pareto tail index of the return distribution. Defaults to 1.25.
- `wrapper_pct` (number): Share of the cohort that is well-packaged but low-substance. Defaults to 0.30.

### `aetre_staking_curve` (~190 tokens)

Sweeps the submission fee across a range and returns the submitter equilibrium at each point, showing how entry volume and low-quality deterrence respond to the staking fee rather than evaluating a single fee.

Input parameters:

- `acceptance_capacity` (integer): Slots available for acceptance. Defaults to 200.
- `api_key` (string): Optional license key.
- `c_gen` (number): Marginal cost of generating a proposal. Defaults to 0.01.
- `max_fee` (number): Highest fee on the swept curve. Defaults to 20.0.
- `private_acceptance_value` (number): Private value of acceptance to the submitter. Defaults to 100.0.
- `steps` (integer): Points on the curve. Defaults to 10, capped at 200.
- `total_potential_applicants` (integer): Size of the applicant pool. Defaults to 5000.

## Diagnostics

Captured diagnostic sections: Provenance. The full working is on the page: https://verifymcp.io/servers/grayclayton-aetre-mcp/ghcr-io-grayclayton-aetre-mcp-0-8-1#diagnostics

## Score history

- 2026-09-21: 36
- 2026-09-20: 36

## Common questions

### What is the AETRE & The Governed Agent MCP server?

AETRE & The Governed Agent is an MCP server listed in the public MCP registry as io.github.grayclayton/aetre-mcp. Decision-theoretic triage and fail-closed agent gating: 32 tools over stdio. This page covers its container image (ghcr.io/grayclayton/aetre-mcp:0.8.1).

### Is the AETRE & The Governed Agent MCP server safe to use?

AETRE & The Governed Agent scores 36 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the AETRE & The Governed Agent MCP server expose?

AETRE & The Governed Agent exposes 33 tools: aetre_system_catalog, aetre_triage_proposal, aetre_calculate_voi, aetre_check_governor, aetre_exploration_audit, and 28 more. Their descriptions and schemas cost roughly 5,289 tokens of context every time the server is loaded.

### Is the AETRE & The Governed Agent MCP server still maintained?

AETRE & The Governed Agent is still listed as active in the MCP registry. We last reached this channel on 21 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Repository: https://github.com/grayclayton/aetre
- Website: https://www.lithiumeel.com/aetre
- Changelog RSS feed: https://verifymcp.io/servers/grayclayton-aetre-mcp/ghcr-io-grayclayton-aetre-mcp-0-8-1.xml
- Changelog JSON feed: https://verifymcp.io/servers/grayclayton-aetre-mcp/ghcr-io-grayclayton-aetre-mcp-0-8-1.json
- HTML version of this page: https://verifymcp.io/servers/grayclayton-aetre-mcp/ghcr-io-grayclayton-aetre-mcp-0-8-1