Claude Code, Skill
Agent Context Kit
Agent Context Kit, a skill listed in the PlugVet directory.
Compatibility: Not stated.
Grade C
One Medium finding. Read the report before installing.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
SECRETS_PATH_CREDENTIAL
Why this grade
- SECRETS_PATH_CREDENTIAL Medium
Mentions a credential store path (~/.ssh, id_rsa, .npmrc, .aws/credentials, credentials.json, .netrc).
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 72
- Key-store/private-key file names such as `.pfx`, `.p12`, `.key`, `id_rsa`, and `id_ed25519` are Critical.
Any High, or two or more Medium findings, give D. One Medium gives C. Low findings or a security Note give B. Each code names a finding behind the letter.
Grades are a hygiene check of the source at the commit shown, on the date checked, not a security audit or a guarantee. The free directory grades are automated. Before release, a PlugVet engineer reviews every paid report and any grade we hold back, including all F grades, against the source code and the evidence we quote.
What was checked
Plain notes from the hygiene checks.
Permissions
Note
requests shell access. The manifest names this permission.
PERMISSION_SHELL_DECLARED
docs/decisions/ADR-0020-github-action-architecture.md, line 46
- Input injection: `args` is parsed with a safe argv splitter (no shell), never passed to `child_process.exec` (only `execFile` with sanitized args). Checked by security fixture (i
docs/decisions/ADR-0021-sdk-and-vscode-integration.md, line 50
- Extension's `extension.ts` runs inside VS Code's Node host and does `import { scanRepository } from "@cynrath/agent-context-kit"` resolved against the installed library (bundled
docs/decisions/ADR-0024-cross-cutting-security-hardening.md, line 58
- `scripts/check-security-boundaries.mjs`: grep forbids `child_process.exec(` (allow `execFile`), `eval(`, `Function(`, `require(userInput`, dynamic `fetch(` in `src/` (only `actio
Show more
docs/rebuild/decisions/ADR-0020-github-action-architecture.md, line 46
- Input injection: `args` is parsed with a safe argv splitter (no shell), never passed to `child_process.exec` (only `execFile` with sanitized args). Checked by security fixture (i
docs/rebuild/decisions/ADR-0021-sdk-and-vscode-integration.md, line 50
- Extension's `extension.ts` runs inside VS Code's Node host and does `import { scanRepository } from "@cynrath/agent-context-kit"` resolved against the installed library (bundled
docs/rebuild/decisions/ADR-0024-cross-cutting-security-hardening.md, line 58
- `scripts/check-security-boundaries.mjs`: grep forbids `child_process.exec(` (allow `execFile`), `eval(`, `Function(`, `require(userInput`, dynamic `fetch(` in `src/` (only `actio
docs/tasks/archive/TASK-0007-v0-2-0-requirements-architecture-baseline-pin-pr.md, line 118
- **security**: grep `NPM_TOKEN|NODE_AUTH_TOKEN` == 0; check no `child_process.exec(` in `src/` (reuse `scripts/check-security-boundaries.mjs` if exists).
docs/tasks/archive/TASK-0008-readiness-context-quality-scoring-engine.md, line 109
- **No arbitrary plugin execution**: no `eval`, `Function`, `require(userInput)`, `child_process.exec` with user content, no downloaded executable packs (`REQ-V020-GOV-007`).
Install and update
Pass
No install-time script found, and dependencies are pinned or absent.
Network
Concern
can reach a host that is not written as a fixed address. can reach example.com. The description does not name that host.
NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST
CHANGELOG.md, line 324
- **Offline-egress invariant** (`scripts/check-offline-egress.mjs`, `tests/security/offline-egress-contract.test.ts`, `tests/security/offline-runtime.test.ts`): static + runtime de
docs/decisions/ADR-0024-cross-cutting-security-hardening.md, line 58
- `scripts/check-security-boundaries.mjs`: grep forbids `child_process.exec(` (allow `execFile`), `eval(`, `Function(`, `require(userInput`, dynamic `fetch(` in `src/` (only `actio
docs/rebuild/decisions/ADR-0024-cross-cutting-security-hardening.md, line 58
- `scripts/check-security-boundaries.mjs`: grep forbids `child_process.exec(` (allow `execFile`), `eval(`, `Function(`, `require(userInput`, dynamic `fetch(` in `src/` (only `actio
Show more
docs/security/THREAT_MODEL.md, line 51
- **Static gate**: `scripts/check-offline-egress.mjs` + `tests/security/offline-egress-contract.test.ts` forbid `fetch`, `node:https`, `http.request/get`, `net.connect`, `tls.conne
docs/tasks/archive/TASK-0008-readiness-context-quality-scoring-engine.md, line 485
- **No network / no exec / no plugin code**: scorer is pure; `grep -R "fetch(|eval(|Function(|child_process.exec(|require(userInput"` over `src/core/readiness/` must be 0. Contract
docs/tasks/archive/TASK-0008-readiness-context-quality-scoring-engine.md, line 550
- `check-security-boundaries` gate - `grep -R "child_process.exec(|eval(|Function(|fetch(" src/core/readiness` == 0; YAML depth/size caps not directly readiness but baseline file s
docs/tasks/archive/TASK-0008-readiness-context-quality-scoring-engine.md, line 599
- Grep gates: `grep -R "fetch(|eval(|child_process.exec" src/core/readiness` → 0 lines (recorded), and `grep -R "REQ-" src/cli/commands/readiness*` → 0 (help leak check).
docs/tasks/archive/TASK-0010-provider-aware-context-profiles.md, line 270
- **Offline guarantee**: grep `fetch(` `http.get` `https.get` in `src/` → 0; network spy test (stub `globalThis.fetch` to throw) → `resolveProfile` still succeeds without calling f
Secrets
Concern
reads a credential or environment file.
SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
.gitignore, line 19
# Local secrets and environment files
.gitignore, line 20
.env
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 72
- Key-store/private-key file names such as `.pfx`, `.p12`, `.key`, `id_rsa`, and `id_ed25519` are Critical.
Show more
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 69
- `.env` remains Critical.
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 70
- `.env.local` or similar non-sample env files remain Critical.
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 71
- `.env.example`, `.env.sample`, `.env.template`, and `.env.dist` are not Critical.
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 118
- Added environment sample precision: `.env.example`, `.env.sample`, `.env.template`, and `.env.dist` are review findings instead of Critical path findings.
docs/tasks/TASK-0012-v020-risk-scanner-precision.md, line 119
- Kept real environment files such as `.env` and `.env.local` Critical.
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Note
the file cap was reached before every file was read.
OBFUSCATION_FILE_CAP_REACHED
docs/tasks/TASK-0240-v100-1-0-0-rc-1-hosted-rc-evidence-execution-and-recording.md, line 1
# TASK-0240: V100 1.0.0-rc.1 hosted RC evidence execution and recording
Licence
Pass
A licence file is present (LICENSE).
Maintainer activity
Pass
The latest commit is within the last 365 days.
Note
Hygiene check of commit 64a9c9d56ea3. requests shell access. The manifest names this permission. can reach a host that is not written as a fixed address. can reach example.com. The description does not name that host. reads a credential or environment file. the file cap was reached before every file was read.
Date and source
Date checked:
github.com
Commit graded: 64a9c9d56ea36eb475de29d050596b506264a07e
This link goes to the original host. PlugVet does not host or check what you download from there, and installing it is your own decision and risk.