Passalong
NPM · PASSALONG · 2 COMPONENTS
One work board for a team and its agents: work, projects, documents and files.
Available components
Last scanned (14 hours ago)
Scores higher than 51% of npm servers
Category Developer Tools
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 Security81
- No malware found by supply-chain analysis.Pass
- CVE check failed: a known high-severity CVE affects zod 3.25.76, a direct dependency. A fixed version is available. View diagnostics → Fail
- No install/post-install scripts declared.Pass
- 31 of 93 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency97
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Cryptographically verified build provenance (signed, bound to codellyson/passalong). View diagnostics → Pass
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 13 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability58
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 4638 tokens (~272/item across 17 items; 17 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 Management24
- Stability check failed: the tool surface changed between 0.13.1 and 1.7.0: 16 tool removals, 0 breaking changes, 12 additions. See how to fix → Fail
Tool Coverage97
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 90% of tool parameters carry a description.Partial
- Structured output schemas are declared (24% of tools); any adoption earns full credit.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- All 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
- An AI judge read all 18 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
Checks we ran (16)
- Malware scan
- Known CVEs
- Install scripts
- Dependency health
- Source repository
- Provenance
- License
- Maintenance
- Security disclosure
- Schema quality
- Stability
- Tool coverage
- Injection markers
- Destructive annotations
- Judged manipulation
- MCP protocol
Does not apply to this package (2)
- Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
- Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
The score does not measure everything. What the score does not measure →
Every 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 Nov 26 +15
- Malware scan: unverified → pass ▲ security
- 18 Nov 26 −27
- Malware scan: pass → unverified ▼ security
- Stability: pass → fail ▼ security
- Tool coverage: 19% → 24% ▲ functional
- Schema quality: 5242 → 4638 ▲ functional
- Schema quality: excellent → good functional
- Stability: pass → 0.97 functional
- Package version: 0.13.1 → 1.7.0 functional
- 12 Nov 26 0
- Stability: 0.97 → pass security
- 11 Nov 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.
- 9 Nov 26 +6
- Source repository: unverified → pass ▲ security
- 8 Nov 26 −5
- Provenance: pass → unverified ▼ security
- 7 Nov 26 −2
- Stability: pass → 0.83 functional
- 1 Nov 26 0
- Stability: 0.97 → pass security
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 21 Nov 2026 · Analysed npm/passalong@1.7.0
Provenance Verified
A signed build attestation was found and verified, binding this exact artifact to the source repository it claims to come from.
| Result | Verified |
|---|---|
| Ecosystem | npm |
| Reason | Verified |
| Discovered via | Registry attestation endpoint |
| Source repo | codellyson/passalong |
| Certificate issuer | https://token.actions.githubusercontent.com |
| Certificate SAN | https://github.com/codellyson/passalong/.github/workflows/release.yml@refs/tags/v1.7.0 |
| Rekor log index | 3746847963 |
| Predicate type | SLSA build provenance https://slsa.dev/provenance/v1 |
| Subject digest | sha512:8e67b19a7e00788fffc24d7dc97f166a36669d0abc9fef891957d983dd654a9e4ba0ef10948072679695d2f8fabefd255e64d9f1c49545272f697ef4c |
Background: How many MCP packages publish verified provenance →
Vulnerabilities 1 finding
| ID | CVE | Severity | Vector | Fix available |
|---|---|---|---|---|
| GHSA-mrhx-c7xc-m3w5 | CVE-2023-54404 | high | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H | yes |
Background: What a vulnerability scan can and cannot prove →
Dependencies 93 packages
| Packages resolved | 93 |
|---|---|
| Stale | 31 |
| Tree resolution | Complete |
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 →
accounts Accounts on this machine ~99
The Passalong accounts signed in on this machine, which one this session is using, and whether it still has to be chosen. With two or more and nothing chosen, every other call is refused until `use` says which: ask the PERSON, in plain words, and pass their answer. It lasts until this session ends and never moves the machine's default.
| Name | Type | Req | Description |
|---|---|---|---|
| use | string | – | an account name from the list, as the person said it |
No output schema declared.
No examples provided.
add_file Add a file ~353
Upload a picture or any other file (a path in `file`, or a picture's bytes in `data` when you have the image and no file) and get back the markdown line that points at it. Files up to 10MB, pictures up to 5MB. WHERE THE LINE GOES DEPENDS ON WHAT THE FILE IS EVIDENCE OF. Proof that work you are handing in holds (a screen that renders right, a total that matches) goes in that check's `ran` on hand_in, and nothing new is written: a check answered with a picture is shown, which is what the hand-in asks for. A picture that is part of the work itself (the screenshot of a bug, a diagram it explains) goes in its text (create_work, or update_work), because work travels as its text and a file beside it does not travel at all. DO NOT WRITE NEW WORK TO CARRY SCREENSHOTS: six images and a caption each is a hand-in. With `project`, the file is kept in that project instead. Pictures (png, jpg, webp, gif) are drawn where they are named; anything else is a file to download.
| Name | Type | Req | Description |
|---|---|---|---|
| data | string | – | a picture itself, base64 (a data: URL is fine), for when a tool handed you the image and never wrote it to disk |
| file | string | – | path to the file on this machine |
| name | string | – | label for it; defaults to its filename |
| project | string | – | keep it in this project instead |
| type | string | – | media type of `data`: image/png, image/jpeg, image/webp or image/gif |
No output schema declared.
No examples provided.
ask Ask the person ~264
You need the person's answer before you can go on. Ask one clear question, with the context it needs to be answered without opening your terminal. You keep what you hold and nobody else can take it. Then STOP and tell the user you are waiting: do not carry on guessing. When they have replied, call take with this id and their reply comes back with it. Use `pass` instead when the work is not yours. When the answer is a choice between options, offer them as `choices` so the person can pick one with a tap.
| Name | Type | Req | Description |
|---|---|---|---|
| choices | array | – | when the answer is a pick between options (two layouts, three fixes): two to four, each with a screenshot when it is something to look at. The person picks one in the hub with a tap; their pick comes… |
| cwd | string | – | the worktree you are working in; default is the server's cwd |
| id | string | yes | passalong id |
| question | string | yes | one clear question in the first person, a few sentences at most, plain text: "Should the dark-mode toggle live in Settings or the header? The header is crowded on mobile." What you were about to do,… |
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | – |
| lease_until | string | – | – |
No examples provided.
create_work Write work down ~223
Write down work someone must still do (one piece, or several at once): a plan broken into steps, or every defect a test run turned up, each its own item so each can be taken on its own. Check what is open first (find_work): if this session answers something you hold, hand_in that instead. Each item is Ready at once and reaches whoever it is for; `draft: true` holds them all for a person to read first. A picture belongs in the text: add_file it and put the line it returns in `body`. Knowledge is not work: what someone needs to know goes into a project's documents (save_document).
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | the repo you are writing from |
| draft | boolean | – | hold it all in Draft for a person to read and make Ready |
| items | array | yes | – |
| to | string | – | team slug, team/@handle for one teammate, team/@ada,@bola for several, or team/#group for the people who do a thing; default is your current team |
No output schema declared.
No examples provided.
delete_document Delete a document ~61
Delete one document, with its earlier versions. It cannot be brought back, so do it only when the person asked for that document to go; to change what it says, use save_document.
| Name | Type | Req | Description |
|---|---|---|---|
| document | string | yes | – |
| project | string | yes | – |
No output schema declared.
No examples provided.
delete_file Delete a file ~51
Delete one file from a project. It cannot be brought back, so do it only when the person asked for that file to go.
| Name | Type | Req | Description |
|---|---|---|---|
| file | string | yes | file id |
| project | string | yes | – |
No output schema declared.
No examples provided.
find_projects Find projects ~173
With `id`, one project: its documents, files and the projects inside it. With `query`, the projects whose name, description or any document mentions those words, each naming its matching documents. With neither, every project you can see, briefly: id, title, repos and how much is in it; ask for one by `id` for the rest. Each answer's `space` says where it lives: private, or a named team; `repo` is the repo it is built from, "" when it has none. Read that repo when you write or check its documents.
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | project id |
| query | string | – | words to look for in projects and their documents |
| scope | string | – | "all" (default), "mine", or a team slug |
No output schema declared.
No examples provided.
find_work Find work ~210
READ work without taking it. With `id`, that piece of work: its markdown, read only; work over 20,000 characters comes back as an outline unless you ask for `section` or `full`. With `query`, the work whose words match. With neither, the board: what needs the person, who is on what, what is ready, and what this worktree holds. To DO the work, call take instead: it is the only way anyone learns somebody picked it up.
| Name | Type | Req | Description |
|---|---|---|---|
| full | boolean | – | with id: the whole text, however long |
| id | string | – | passalong id (e.g. k3mq2xa7) or share link |
| query | string | – | words to look for in titles, tags and text |
| scope | string | – | with query: "mine", "all" (default), or a team slug |
| section | string | – | with id: only the part under this heading; see the outline |
No output schema declared.
No examples provided.
hand_in Hand it in ~630
Finished here, with proof. IT IS FINAL: the work goes to Done with what you showed, and nobody reviews it after you (anyone can reopen it with a reason). So SHOW IT: a screenshot for anything with a screen (before and after, for a defect) or, for backend work, the one command and the line of its output that shows it works. Send `checks`: one entry per thing the work asked for, each with that line and the evidence for it. That is what a person reads, line against line, AND IT IS HOW YOU KNOW YOU ARE DONE: every line has one. Give a check `cmd` when a command proves it: the command is run here before the hand-in lands, a non-zero exit refuses it, and its exit code, not your account of it, decides the check. Use `ran` for a check no command can settle: add_file the screenshot and put the line it returns there. `evidence` as one block works too. DO NOT write new work to carry your evidence: it belongs on this call. What you had to adapt goes in `writeup`; what you learned about the project goes into its documents (save_document). Pushed a branch or opened a pull request? Send its link as `pr`: it is how a person sees what changed.
| Name | Type | Req | Description |
|---|---|---|---|
| checks | array | – | one entry per thing the work asked for, in the order you worked them. Each is RUN (`cmd`) or SHOWN (an add_file line in `ran`). Pasting a command you typed into `ran` is not running it and is refused |
| cwd | string | – | the worktree you are working in; default is the server's cwd |
| evidence | string | – | what you ran and what came back, with the command and the lines that decided it, a test summary, a link to the change, or a screenshot url. "it works" is a claim, not evidence |
| id | string | yes | passalong id |
| note | string | yes | one plain sentence, in the first person, saying what you did and how it went: "Added the toggle to Settings; it survives a reload." It is the first thing the person reads, and the evidence is behind… |
| pr | string | – | the web link to the pull request, branch or commit with this change, when you pushed one: the person opens it to see what changed. Left out, the open pull request for this branch is used, if there is… |
| risk | string | – | what this could break, in one line: a shared helper changed, a migration that cannot be undone, a route other clients call. Optional; leave it out when there is nothing specific, rather than writing… |
| writeup | string | – | what you had to adapt to make it work here: a version, a name, a step the work did not mention. 500 characters at most; anything longer belongs in the change. Optional: leave it out when it went as w… |
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | – |
| state | string | – | – |
No examples provided.
pass Pass it ~105
Not yours to do, or you are stuck: give it back with the reason. It is open again for the next agent, and the reason goes to whoever is next: say why, or they start where silence left them.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | the worktree you are working in; default is the server's cwd |
| id | string | yes | passalong id |
| why | string | yes | one line: why it is not yours, or where you got stuck |
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | – |
| passed | boolean | – | – |
No examples provided.
progress Report progress ~128
Say you are still on what you hold, with a one-line note the hub shows. 30 minutes without one marks it stalled. If the answer says you no longer hold it, stop.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | the worktree you are working in; default is the server's cwd |
| id | string | yes | the id of what you hold |
| note | string | – | one plain sentence in the first person, as you would tell the person: "Reading the settings page to find where the toggle goes." Start it with BLOCKED: when you are stopped and need an answer |
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | – |
| lease_until | string | – | – |
| note | string | – | – |
No examples provided.
read_document Read a document ~96
One document's Markdown and its `version`, which save_document needs. `history: true` lists its earlier versions, newest first; `version` returns that version's text. To bring one back, save it over the current version.
| Name | Type | Req | Description |
|---|---|---|---|
| document | string | yes | document id |
| history | boolean | – | list its earlier versions |
| project | string | yes | project id |
| version | integer | – | read this earlier version |
No output schema declared.
No examples provided.
read_file Read a file ~48
Read a picture or a small text file kept in a project into this conversation. Other files stay downloads in the hub.
| Name | Type | Req | Description |
|---|---|---|---|
| file | string | yes | file id |
| project | string | yes | – |
No output schema declared.
No examples provided.
save_document Save a document ~300
Write what someone needs to know into a project. With no `document`, a new one: `name` and `body`. With `document`, change it: `body` with the `version` read_document gave you (a newer edit is refused, so nobody's writing is lost) and/or `name` to rename it, and/or `move_to` another project in the same space. Its id and history stay. Update the page that already covers something rather than adding another. A document has no owner and no state and tells nobody: anything someone must still act on is work (create_work). The README is special: an agent taking work in the project is shown its first 4,000 characters. Put there what it must know before changing anything (what this is, the rules not to break, the words used, where things are) and the detail on other pages, named in it. Refer to other pages by their names, not file names.
| Name | Type | Req | Description |
|---|---|---|---|
| body | string | – | the whole Markdown text |
| document | string | – | the document to change; leave out for a new one |
| move_to | string | – | a project id in the same space to move it into |
| name | string | – | its name: required for a new one, a rename otherwise |
| project | string | yes | project id |
| version | integer | – | with `body` on an existing document: the version you read |
No output schema declared.
No examples provided.
save_project Save a project ~324
With no `id`, make a project, after find_projects shows none fits: a `title` and a short `description`, and its `repo` when it has one, or `repos` when it is built from several (a web app, its API, its mobile app: one product, one project). Leave out `team` for a private one, and pass a team slug only when the person wants it shared; `parent` puts it inside another, in that one's space. With `id`, change its title, description, repo ("" for none), repos (the whole list, the main one first), colour or place (`parent` "" for the top level); its space never changes. The answer's `space` says where it is: tell the person.
| Name | Type | Req | Description |
|---|---|---|---|
| color | string | – | – |
| description | string | – | what it is for, in a line |
| id | string | – | the project to change; leave out to make one |
| parent | string | – | project to put it inside; "" for the top level |
| repo | string | – | the repo it is built from, e.g. github.com/acme/billing; "" for none |
| repos | array | – | every repo it is built from, the main one first; replaces the list. Work written in any of them is filed in this project and its agents read this project's README |
| team | string | – | new project only: team slug; omit for private |
| title | string | – | name of the product or body of work |
No output schema declared.
No examples provided.
take Take work ~170
Say you are doing it, and get it. With `id`, that piece of work. With no id, the next thing waiting here: always something new, never what you hold. While you hold it no other agent can take it, and whoever sent it sees you are on it. You may hold several; the person watches them on the board. If somebody else has it, the answer says who: tell the user rather than doing the work twice. Every answer ends with what to call next.
| Name | Type | Req | Description |
|---|---|---|---|
| any | boolean | – | work for another repo, or none: only when the user asks |
| cwd | string | – | the worktree you are working in; default is the server's cwd |
| id | string | – | passalong id or share link; leave out for the next one |
No output schema declared.
No examples provided.
update_work Change work ~200
Change work that exists, when the person asks or when you wrote it. `status`: ready (whoever it is for may take it), draft (held for a person to read), or archived (off the board; it comes back to Draft). `to`: who it is for: @handle, several as @ada,@bola, #group, or "" for the whole team; whoever held it and is left out has it taken back and is told. `markdown`: its whole text, frontmatter first, as find_work gave it: say what changed, keep what did not.
| Name | Type | Req | Description |
|---|---|---|---|
| cwd | string | – | the repo you are working in |
| id | string | yes | passalong id |
| markdown | string | – | the whole document, frontmatter first, to replace its text |
| status | string | – | – |
| to | string | – | "@handle", "@ada,@bola", "#group", or "" for the team |
No output schema declared.
No examples provided.
How do I install the Passalong MCP server?
Passalong runs locally as an npm package, launched with npx -y passalong mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · passalong
claude mcp add codellyson-passalong -- npx -y passalong mcp
What is the Passalong MCP server?
Passalong is an MCP server listed in the public MCP registry as io.github.codellyson/passalong. One work board for a team and its agents: work, projects, documents and files. This page covers its npm package (passalong).
Is the Passalong MCP server safe to use?
Passalong scores 75 out of 100 on VerifyMCP. We recorded 1 known advisory against it as of 21 November 2026. It declares no install or post-install scripts. Its build provenance is signed and verified. 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 Passalong MCP server expose?
Passalong exposes 17 tools: take, progress, hand_in, pass, ask, and 12 more. Their descriptions and schemas cost roughly 3,435 tokens of context every time the server is loaded.
Is the Passalong MCP server still maintained?
Passalong is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the Passalong MCP server under?
Passalong declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.
What did VerifyMCP test on the Passalong MCP server?
VerifyMCP ran 16 checks on Passalong, covering Malware scan, Known CVEs, Install scripts, Dependency health and 12 more. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.