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 Detective-XH/DocGraph --skill assessment-drift-auditgit clone --depth 1 https://github.com/Detective-XH/DocGraphWrote 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/detective-xh/docgraph/assessment-drift-audit)<a href="https://agentmods.dev/skills/detective-xh/docgraph/assessment-drift-audit"><img src="https://agentmods.dev/badge/skills/detective-xh/docgraph/assessment-drift-audit/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/detective-xh/docgraph/assessment-drift-audit"><img src="https://agentmods.dev/badge/skills/detective-xh/docgraph/assessment-drift-audit.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 | $0.00055 | $0.00864 |
| Opus 5 | $0.00028 | $0.00432 |
| Sonnet 5 | $0.00011 | $0.00173 |
| Haiku 4.5 | $0.00006 | $0.00086 |
Grade A, and why
assessment-drift-audit 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Assessment/Research Drift Audit
Invocation — Claude Code:
Skill("assessment-drift-audit")| OpenCode / Codex: use triggers above.
Runs the DocGraph assessment drift audit and presents findings for human review. The audit is advisory — findings highlight candidates for action, not authoritative rulings.
Execution
- Model: Haiku (query) → Sonnet (remediation suggestions if requested)
- Effort: Low
- Advisor: Not required
- Core computation: done by DocGraph's
GetDriftFindingsengine, not by this skill
Step 1: Check index status
docgraph_status
Confirm:
assessment_driftpack is listed under Domain Packs- Research metadata is indexed (docs with research metadata > 0)
If assessment_drift is not listed, the binary is too old or the pack was never registered —
tell the user to upgrade DocGraph and run docgraph index --force.
Step 2: Run the drift audit
docgraph_context task="research/assessment drift audit" format=drift_audit
This calls DocGraph's GetDriftFindings engine, which checks:
| Code | Meaning |
|---|---|
research.stale_assessment |
valid_until date has passed; assessment is expired |
research.unverified_evidence |
last_verified date is older than 180 days (configurable) |
research.competing_interpretations |
Multiple documents share the same claim_id but have conflicting confidence or analyst_status |
research.superseded_claim |
A research document with superseded_by set is still referenced by other active documents |
research.impacted_deliverable |
A deliverable (deliverable_id set) is linked to an expired assessment |
Step 3: Present findings to the user
Group findings by severity (warnings first, then info).
For each finding:
- State the file path(s) and finding code
- Show the message and evidence from the finding
- Suggest a concrete remediation (below) — but do NOT make the change without confirmation
Remediation patterns
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 · 105 lines · 55 tokens per session scan A 9e3d1c755124
assessment-drift-audit is a skill published in the GitHub repository Detective-XH/DocGraph (8 stars, last pushed 2d ago), licensed MIT. It adds 55 tokens to every session and 864 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 2026-08-31.
Other skills, from other repositories
ori-memory
Persistent agent memory with learning retrieval. Knowledge graph on markdown files — capture insights, decisions, research, and learnings during work, then retrieve them weeks or months later. Use when knowledge is too valuable to lose but too much to inject into every prompt.
okf-loom
Load this repo skill when working with Open Knowledge Format (OKF) bundles: validate, search, discover, update, render, serve the live studio, author concepts, or use the bundled OKF docs. The repo is the distribution unit; run scripts from the checkout with scripts/okf-loom.
tela-wiki
Use when the user asks something their team's own documentation would answer, or asks to write or update a page, spreadsheet, or slide deck in their tela wiki.
document-as-you-go
Quando você termina uma tarefa/commit, pague a dívida de documentação da wiki antes de fechar. Use este skill no Claude Code, OpenCode, Codex CLI ou qualquer agente que tenha a skill MCP do livewiki.
bootstrap-wiki
Bootstrap a repository's initial livewiki documentation through MCP when no wiki pages exist or the user explicitly asks for its first full wiki. Use for deliberate, long-running initial documentation.
context-atlas
Retrieve and interpret durable, evidence-backed project memory from Context Atlas, including overviews, architecture, chronology, decisions, risks, provenance, health, and task-specific context packs. Use when Codex must onboard to an unfamiliar repository, explain how or why code evolved, prepare context before a…