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 agentmods add skills/eliottreich/taskbounty-check/taskbounty-securitynpx skills add eliottreich/taskbounty-check --skill taskbounty-securitygit clone --depth 1 https://github.com/eliottreich/taskbounty-checkWrote 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/eliottreich/taskbounty-check/taskbounty-security)<a href="https://agentmods.dev/skills/eliottreich/taskbounty-check/taskbounty-security"><img src="https://agentmods.dev/badge/skills/eliottreich/taskbounty-check/taskbounty-security.svg" alt="Measured on agentmods" 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 | $0.00094 | $0.01028 |
| Opus 5 | $0.00047 | $0.00514 |
| Sonnet 5 | $0.00019 | $0.00206 |
| Haiku 4.5 | $0.00009 | $0.00103 |
Grade A, and why
taskbounty-security 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 6d 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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TaskBounty security check (GitHub Actions / CI hygiene)
taskbounty-check is a local, zero-dependency checker for GitHub Actions and CI maintenance hygiene:
third-party action pinning, workflow token permissions, and dependency-update automation. It reads
workflow files on disk. By default it makes no outbound network requests and uploads nothing.
Scope, stated honestly. This covers GitHub Actions, CI workflow permissions, action pinning, and update automation. It is not a complete application-security audit - it does not check exposed secrets, authentication, payments, webhooks, or runtime behavior. Say so when you report results.
Hard rules for the agent
- Never upload the user's source code or scan results anywhere. The tool keeps everything local; keep it that way.
- Never modify files without explicit user approval. Findings and fix plans are proposals.
- Never commit, push, open pull requests, or post comments automatically. The user does that.
- Do not add the
--gh-orgflag unless the user explicitly asks to scan an organization (it is the only mode that uses the network, via the user's ownghsession).
1. Run the checker locally (primary)
npx -y [email protected] .
This writes a local report and prints a summary. No network, nothing uploaded. Run it from the repo
root (or pass a path). To see what it would do without writing files, add --dry-run.
2. Interpret the findings
Each finding has a rule id and a confidence label (confirmed vs review). Common rules:
- unpinned-action - a third-party action uses a movable tag/branch (e.g.
@v4) instead of a full commit SHA. Movable refs can be re-pointed upstream. - broad-permissions / no-permissions-block - the workflow grants (or defaults to) broad
GITHUB_TOKENpermissions instead of least privilege. - prt-checkout-untrusted / secrets-in-prt - a
pull_request_targetworkflow runs untrusted PR code and/or exposes secrets to it. - script-injection - untrusted
${{ github.event.* }}input is interpolated into a run step.
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.
- 6d ago First seen · 90 lines · 94 tokens per session scan A 377840af159b
taskbounty-security is a skill published in the GitHub repository eliottreich/taskbounty-check (2 stars, last pushed 2mo ago), licensed MIT. It adds 94 tokens to every session and 1,028 once invoked, about $0.0005 per session on Opus 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
securing-github-actions-workflows
This skill covers hardening GitHub Actions workflows against supply chain attacks, credential theft, and privilege escalation. It addresses pinning actions to SHA digests, minimizing GITHUBTOKEN permissions, protecting secrets from exfiltration, preventing script injection in workflow expressions, and implementing…
securing-github-actions-workflows
This skill covers hardening GitHub Actions workflows against supply chain attacks, credential theft, and privilege escalation. It addresses pinning actions to SHA digests, minimizing GITHUBTOKEN permissions, protecting secrets from exfiltration, preventing script injection in workflow expressions, and implementing…
GitHub Actions Workflow Security Review
Reviews GitHub Actions workflow diffs for script injection of untrusted input, pullrequesttarget with untrusted checkout, over-broad GITHUBTOKEN permissions, and unpinned third-party actions.
github-workflow-automation
Advanced GitHub Actions workflow automation with AI swarm coordination, intelligent CI/CD pipelines, and comprehensive repository management.
ci-triage
Classify CI failures — distinguish clear regressions from infra flakes and security-test failures. Produces structured failure reports.
publish
Publish the Octop Python package: cut a release branch from develop, bump version, update CHANGELOG, open a PR to main; after merge, Actions tag on main (PyPI / Docker Hub + GHCR) and sync main into develop. Use when the user asks to publish, release, bump version, cut a release, or run /publish.