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 squerne/open-career-skills --skill evidence-minergit clone --depth 1 https://github.com/squerne/open-career-skillsWrote 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/squerne/open-career-skills/evidence-miner)<a href="https://agentmods.dev/skills/squerne/open-career-skills/evidence-miner"><img src="https://agentmods.dev/badge/skills/squerne/open-career-skills/evidence-miner/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/squerne/open-career-skills/evidence-miner"><img src="https://agentmods.dev/badge/skills/squerne/open-career-skills/evidence-miner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 47 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00067 | $0.00963 |
| Opus 5 | $0.00034 | $0.00481 |
| Sonnet 5 | $0.00013 | $0.00193 |
| Haiku 4.5 | $0.00007 | $0.00096 |
Grade A, and why
evidence-miner 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 12d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evidence Miner
The other skills in this workspace refuse to invent facts. This one goes further: it harvests real ones. People systematically forget their own achievements; their git history doesn't. (Concept inspired by Play-New/apply-new, MIT: career evidence should come from work artifacts, and every claim should trace to data.)
Privacy contract (state it to the user up front, once)
Everything here is read-only and stays local: repo content is read on this machine and goes nowhere except into story files the user approves. Never run write/network git commands. If a repo involves a client or employer the user may not want named, offer to redact the name in the filed story ("a fintech client" instead of the name).
Step 1: Scope
Ask the user which repo(s) or work folders to mine (absolute paths), and roughly what period matters. If they own PRs on GitHub, gh widens the evidence.
Step 2: Harvest (bash, read-only, no permission theater)
Use the bash tool to execute read-only git and gh commands directly. Do not ask for permission for read-only git log/git shortlog/git show --stat/gh pr list commands; execute them and gather the evidence. Never execute anything that writes (no checkout, commit, push, config).
Per repo, run what the situation needs, typically:
git -C <repo> log --author="<user>" --oneline --since="12 months ago"
git -C <repo> shortlog -sn --since="12 months ago" # their share of the work
git -C <repo> log --author="<user>" --stat --since="12 months ago" | head -400
gh pr list --repo <owner/repo> --author "@me" --state merged --limit 50 --json title,mergedAt,additions,deletions # when gh is available
Also skim CHANGELOGs, ADRs, or docs folders the user points at.
Step 3: Detect achievement signals
Look for clusters, not single commits: a shipped feature (branch/PR series landing in one area), a performance or cost fix (commit messages with numbers: "cut build from 12m to 3m"), an incident or bug saga resolved, a migration or refactor completed, sustained ownership of a subsystem (shortlog dominance), tooling or process the team adopted.
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.
- 12d ago First seen · 52 lines · 67 tokens per session scan A a80af67eae77
evidence-miner is a skill published in the GitHub repository squerne/open-career-skills (22 stars, last pushed 1mo ago), licensed MIT. It adds 67 tokens to every session and 963 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
docs
Public-facing skill docs. Full mode prompts live in .agents/skills/careerops/.
careerops
CareerOps agent skill — open-source career OS modes (scan, evaluate, rank, tailor, interview, followup, outcome, advise). Materials-first. Never invents experience. Never auto-applies.
mem0-test-integration
Verify a Mem0 integration produced by /mem0-integrate. Runs in the same workspace on the same branch (loose coupling) — installs dependencies, runs the repo's native test suite, then exercises a real end-to-end smoke flow against the user's API key. Produces a scorecard. TRIGGER when: user has just run /mem0-integrate…
baby-sit
Monitor a GitHub pull request until CI is green, diagnose failures, and rerun only evidence-backed flaky GitHub Actions jobs.
git-checkpoint
Use git as a safety net - create a checkpoint commit before risky or large changes and roll back cleanly if a change makes things worse. Use before multi-file refactors.
meta-codereview-current-diff
Read the current uncommitted diff, run three independent reviewers (safety + tests-coverage + style) in parallel, then arbitrate a single BLOCK / BLOCKWITHOVERRIDE / PASSWITHNOTES verdict. Use before commit when you want a multi-perspective second-opinion instead of a single-reviewer agent loop.