io.github.derekchoyai/hone
MCPB · HONE.MCPB · SCANNED SEP 20
Score your judgment over AI-assisted work. Hone scores you, not the model.
Available components
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. How we score → Why this is hard to score →
Supply Chain Security23
- Malware scan not yet available for this package.Unverified
- Known CVEs were checked across the 96 of 97 dependencies we could resolve, so this covers what we could see, not the whole tree.Partial
- No install/post-install scripts declared.Pass
- Dependency health was assessed across the 96 of 97 dependencies we could resolve, so this covers what we could see, not the whole tree. View diagnostics → Partial
Provenance & Transparency48
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (Apache-2.0).Pass
- Actively maintained (last published 91 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability59
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 991 tokens (~247/item across 4 items; 4 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management93
- Stability observed for 28 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage86
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 59% of tool parameters carry a description.Partial
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 4 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 4 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Add this component to your MCP client. Where a client-specific snippet is available, pick your client below and copy it straight into your config; otherwise use the connection detail shown.
mcpb · hone.mcpb
Download bundleEvery change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 19 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 87 to 90. That category is still filling its 30-day observation window: 26 days of observed history at the previous scan, 27 at this one. The score rises as the window fills, whether or not the server changes.
- 17 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 80 to 83. That category is still filling its 30-day observation window: 24 days of observed history at the previous scan, 25 at this one. The score rises as the window fills, whether or not the server changes.
- 16 Sept 26 −3
- Stability: pass → 0.80 functional
- 15 Sept 26 0
- Stability: 0.97 → pass security
- 14 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 93 to 97. That category is still filling its 30-day observation window: 28 days of observed history at the previous scan, 29 at this one. The score rises as the window fills, whether or not the server changes.
- 12 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 87 to 90. That category is still filling its 30-day observation window: 26 days of observed history at the previous scan, 27 at this one. The score rises as the window fills, whether or not the server changes.
- 10 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 80 to 83. That category is still filling its 30-day observation window: 24 days of observed history at the previous scan, 25 at this one. The score rises as the window fills, whether or not the server changes.
- 9 Sept 26 −3
- Stability: pass → 0.80 functional
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 20 Sept 2026 · Analysed mcpb/https://github.com/derekchoyai/hone/releases/download/v0.3.0/hone.mcpb@0.3.0
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | mcpb |
Background: How many MCP packages publish verified provenance →
Dependencies 96 packages
| Packages resolved | 96 |
|---|---|
| Stale | 31 |
| Tree resolution | Partial |
The dependency tree was only partially resolved, so these counts may be incomplete.
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
get_profile ~92
The user's longitudinal AI-Q profile from this machine's local history (~/.hone/profile.json): review count, per-dimension Discernment averages, weakest/strongest dimension, composite and calibration trends, Delegation stats when assessed, qualitative Design indicators, and recent reviews. Use it to answer 'how is my judgment developing?' or to ground coaching in their actual history. Contains judgment metadata only — never any reviewed work.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_protocol ~279
Get the open AI-Q standard protocol for computing the user's AI-Q (their judgment over AI-assisted work — the three Ds: Discernment, Delegation, Design) in YOUR context. mode='after' (default) reviews FINISHED AI-assisted work: dimension definitions, domain rubric, interview rules (incl. the delegation probe), question count, techniques, deterministic scoring, and the report + teaching (3-5 lessons) structure. mode='during' is THINK-FIRST — Delegation coaching: the user has not produced anything yet; sharpen their prompt and prime their judgment BEFORE the AI answers (no Discernment score exists yet). You (the host model) already have the work/task loaded — you do the reading and run the interview; this tool supplies the methodology. Call this BEFORE interviewing the human.
| Name | Type | Req | Description |
|---|---|---|---|
| context | string | – | Who/where: adapts language, tone, depth, and the scoring bar. |
| domain | string | yes | The kind of work being reviewed (or the task domain in think-first mode). |
| mode | string | – | 'after' = review finished work (default, scored). 'during' = think-first: prime judgment before using AI (not scored). |
| riskLevel | string | – | low = throwaway; medium = shared/internal; high = production, money, health, legal, irreversible. |
No output schema declared.
No examples provided.
record_review ~335
Save a completed judgment review to the user's LOCAL profile (~/.hone/profile.json) — judgment METADATA ONLY (scores, domain, a short subject label), never the work itself. Pass the six Discernment dimensions, plus the three Delegation facets when assessed; a think-first delegation read passes just the facets. All sub-scores and the AI-Q composite are recomputed server-side (passed composites are ignored — anti-gaming). Future get_protocol calls personalize from this history. Returns the updated trend and an encouragement read. Tell the user you're saving scores only, and skip this call if they decline.
| Name | Type | Req | Description |
|---|---|---|---|
| accountability | – | – | – |
| assumptionAwareness | – | – | – |
| briefQuality | – | – | 1-5 — Delegation facet; pass ONLY when assessed behind the evidence gate. |
| confidenceCalibration | – | – | – |
| context | string | – | – |
| domain | string | yes | The kind of work that was reviewed. |
| iterationControl | – | – | 1-5 — Delegation facet; pass ONLY when assessed. |
| riskLevel | string | – | – |
| riskRecognition | – | – | – |
| statedConfidence | number | – | The 0-100 confidence the human stated BEFORE the reveal. |
| subject | string | – | ≤8-word neutral label of WHAT was reviewed (e.g. 'Q3 GTM strategy deck') — a recall anchor, never the content. |
| taskSelection | – | – | 1-5 — Delegation facet; pass ONLY when assessed. |
| topGap | string | – | One short sentence: the single most important gap from this review. |
| understanding | number | – | 1-5 |
| verification | – | – | – |
No output schema declared.
No examples provided.
score_review ~285
Deterministically score a completed AI-Q judgment interview into the user's AI-Q. Pass the six 1-5 Discernment dimension scores (judged by comparing the human's answers to your private work map), and — ONLY if your interview surfaced the brief/iteration history — the three 1-5 Delegation facets. Returns the official AI-Q composite (0-100, weights renormalized over what was assessed), band, sub-scores, and calibration read. For a think-first delegation-only read, pass just the three facets. ALWAYS use this instead of computing any score yourself.
| Name | Type | Req | Description |
|---|---|---|---|
| accountability | – | – | – |
| assumptionAwareness | – | – | – |
| briefQuality | – | – | 1-5 — Delegation: did the brief carry the context, constraints, and success criteria the task needed? Pass ONLY with evidence from the interview. |
| confidenceCalibration | – | – | – |
| iterationControl | – | – | 1-5 — Delegation: caught and corrected drift between first output and final. Pass ONLY with evidence. |
| riskRecognition | – | – | – |
| statedConfidence | number | – | The 0-100 confidence the human stated BEFORE seeing any results. |
| taskSelection | – | – | 1-5 — Delegation: right task to hand to AI; knew what to keep human. Pass ONLY with evidence. |
| understanding | number | – | 1-5 |
| verification | – | – | – |
No output schema declared.
No examples provided.
What is the io.github.derekchoyai/hone MCP server?
io.github.derekchoyai/hone is an MCP server listed in the public MCP registry as io.github.derekchoyai/hone. Score your judgment over AI-assisted work. Hone scores you, not the model. This page covers its MCPB bundle (https://github.com/derekchoyai/hone/releases/download/v0.3.0/hone.mcpb).
Is the io.github.derekchoyai/hone MCP server safe to use?
io.github.derekchoyai/hone scores 57 out of 100 on VerifyMCP. It declares no install or post-install scripts. 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 io.github.derekchoyai/hone MCP server expose?
io.github.derekchoyai/hone exposes 4 tools: get_protocol, score_review, record_review, get_profile. Their descriptions and schemas cost roughly 991 tokens of context every time the server is loaded.
What licence is the io.github.derekchoyai/hone MCP server under?
io.github.derekchoyai/hone declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.