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 me-cedric/agent-compass --skill vuln-triagegit clone --depth 1 https://github.com/me-cedric/agent-compassWrote 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/me-cedric/agent-compass/vuln-triage)<a href="https://agentmods.dev/skills/me-cedric/agent-compass/vuln-triage"><img src="https://agentmods.dev/badge/skills/me-cedric/agent-compass/vuln-triage/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/me-cedric/agent-compass/vuln-triage"><img src="https://agentmods.dev/badge/skills/me-cedric/agent-compass/vuln-triage.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 | $0.00060 | $0.00908 |
| Opus 5.5 | $0.00024 | $0.00363 |
| Sonnet 5.5 | $0.00012 | $0.00182 |
| Haiku 4.5 | $0.00006 | $0.00091 |
Grade A, and why
vuln-triage 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 10d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vuln-triage
Rank the findings first. Fix the highest score first. Never suppress a finding without a reason and a date.
Collect
- osv-scanner:
osv-scanner scan source --recursive --config .osv-scanner.toml --format sarif --output-file osv.sarif .- Exit 1 means "findings". It is not a failure of the scanner. Exit 127 or 128 is a failure.
- Offline: add
--offline-vulnerabilities --download-offline-databases. SetOSV_SCANNER_LOCAL_DB_CACHE_DIRECTORYto keep the database between runs.
- Checkmarx One:
cx results show --scan-id <id> --report-format sarif --output-name cx --output-path .. Do not start a scan or create a project. Ask the user first. - SonarQube has no SARIF export. Use the SonarQube MCP (read-only) to list the issues of type
VULNERABILITYand the security hotspots. - Merge the SARIF files and apply the gate:
node .agents/skills/vuln-triage/scripts/sarif-merge.mjs --out merged.sarif --fail-at 9 osv.sarif cx.sarif
The script keeps one run per tool and drops duplicates. The fingerprint is the rule id, the path and a hash of the snippet,
so a finding keeps its fingerprint when its line moves. Exit 1 lists each finding at or above the --fail-at score.
Rank
| CVSS | Class | Action |
|---|---|---|
| 9.0 or more | critical | Fix now. The CI gate fails. |
| 7.0 to 8.9 | high | Fix in this sprint. |
| 4.0 to 6.9 | medium | Make a ticket. |
| less than 4.0 | low | Fix in a batch upgrade. |
Then adjust the order: a finding in a dev-only dependency or in code that no route reaches goes down. A finding with a known exploit goes up.
Fix
- Dependency: upgrade to the first fixed version that osv lists. For a transitive dependency, upgrade the parent first. If no parent release has the fix, add an override (
pnpm.overrides, npmoverrides, yarnresolutions). - Code: make the smallest fix. Add a test that fails before the fix.
- Install from the lockfile and run the tests.
- Scan again. The finding must be gone.
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.
- 10d ago First seen · 67 lines · 60 tokens per session scan A b92092cb1e19
vuln-triage is a skill published in the GitHub repository me-cedric/agent-compass (3 stars, last pushed 4d ago), licensed MIT. It adds 60 tokens to every session and 908 once invoked, about $0.0002 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 2026-10-18.
Other skills, from other repositories
worker-integration
Worker-Agent integration for intelligent task dispatch and performance tracking.
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
jetson-diagnostic
Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.