Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add paultyng/skill-issue --skill evaluate-dependencygit clone --depth 1 https://github.com/paultyng/skill-issueWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/paultyng/skill-issue/evaluate-dependency)<a href="https://agentmods.dev/skills/paultyng/skill-issue/evaluate-dependency"><img src="https://agentmods.dev/badge/skills/paultyng/skill-issue/evaluate-dependency/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/paultyng/skill-issue/evaluate-dependency"><img src="https://agentmods.dev/badge/skills/paultyng/skill-issue/evaluate-dependency.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00149 | $0.03081 |
| Opus 5.5 | $0.00060 | $0.01232 |
| Sonnet 5.5 | $0.00030 | $0.00616 |
| Haiku 4.5 | $0.00015 | $0.00308 |
Grade A, and why
evaluate-dependency scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evaluate Dependency
Dual-purpose: upfront selection when adding a new dep, and review when a PR adds or bumps one. Same evaluation criteria either way, with per-language addenda for ecosystem-specific quirks.
When to use
| Mode | Triggers |
|---|---|
| Selection | Considering adding a dep, asks "should I use X", compares alternatives, before go get / npm install / pip install / cargo add |
| Review | PR diff touches a manifest, or review-security / review-code sees a new dep in scope |
Language-specific quirks (read first)
The canonical package coordinate usually has ecosystem-specific gotchas that override the general criteria. The wrong coordinate is the wrong dep, no matter how good the package is.
- Go: references/go.md. Semantic import versioning (
/v2,/v3paths),+incompatiblesmell,pkg.go.dev/vuln.go.dev,govulncheck. - Other ecosystems: not yet covered here. Run the general checklist and flag uncertainty about the canonical coordinate with
unsure:(perterse-output) rather than guessing.
Future addenda: npm (scoped vs unscoped, deprecated packages, ESM/CJS), Python (package vs distribution name, wheel/sdist, ABI compatibility), Rust (pre-1.0 churn, crate renames), Ruby (transitive native-extension hazards). Add when a real evaluation surfaces the need.
Evaluation criteria
Run through these. Cite specific data: no hand-waving, no I think.
-
License. Detect via the ecosystem's registry or the repo's
LICENSE/LICENSE.md/COPYING. Permissive (MIT, BSD, Apache-2.0, ISC, MPL-2.0) is generally fine. Copyleft (GPL, AGPL, LGPL) needs an explicit compatibility check against the user's project license. No license = no use. Note relicense history; relicenses are a red flag. -
Internal precedent. Before weighing external popularity, check whether the candidate (or an equivalent) is already used or wrapped internally. Sources, in order:
- Project-local hints:
CLAUDE.md,REVIEW.md,.claude/rules/, and the manifests of sibling services if mounted locally. - Internal code search if available: Sourcegraph (
mcp__sourcegraph__keyword_search/mcp__sourcegraph__list_repos),gh search codescoped to the user's orgs. - Internal package registry / module index if the project documents one.
- Project-local hints:
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 121 lines · 149 tokens per session scan A 387bcdca675e
evaluate-dependency is a skill published in the GitHub repository paultyng/skill-issue (8 stars, last pushed yesterday), licensed MIT. It adds 149 tokens to every session and 3,081 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
chatcrystal-debug-recall
Recall ChatCrystal memories for debugging tasks involving failing tests, compiler errors, runtime exceptions, dependency issues, environment breakage, or performance regressions. Use when historical root causes, fixes, or pitfalls may accelerate diagnosis before proposing a fix.
verify
Fact-check claims encountered during reading — dates, names, events, citations. Use when encountering historical facts or disputed claims.
megalinter-check
Collect MegaLinter lint errors for the current repository. Use when the user wants to know if the code passes linting, why the MegaLinter CI job fails, or before/after fixing lint errors. Two modes - watch a CI job (GitHub Actions, GitLab CI, Azure Pipelines, Bitbucket Pipelines) and parse its logs, or run MegaLinter…
assistant
Assistant — on any repo, scan README→docs→AGENTS→CONTRIBUTING→PR templates→task runners→devcontainer→CI→configs before code; cite sources; prefer AGENTS.md for agent behavior; portable across Cursor/Copilot/Claude; use agent-toolkit CLI when needed.
design-improvement
WHAT - Browser-grounded iterative design improvement. Consumes design-assessment findings, defines direction, prioritizes safe vs ambiguous changes, implements within existing design system, runs app, captures rendered evidence via browser, reviews and iterates. Reuses evidence model — no new scoring framework.
codeql
CodeQL operational workflow — discover/config, run/inspect, triage SARIF findings (rule/query ID, source→sink, evidence), remediate, re-validate. Distinguishes broad MegaLinter linting from semantic security analysis.