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 commands/not-diamond/self-care/validategit clone --depth 1 https://github.com/Not-Diamond/self-careWhat 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 | $0.00014 | $0.00946 |
| Opus 5 | $0.00007 | $0.00473 |
| Sonnet 5 | $0.00003 | $0.00189 |
| Haiku 4.5 | $0.00001 | $0.00095 |
Grade A, and why
validate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Validate Trace
Validate the provided trace file and report its metadata. Supports both OTEL (JSON) and Claude Code (JSONL) formats.
Instructions
Run the deterministic trace validator:
node "${CLAUDE_PLUGIN_ROOT}/agents/tools/validate-trace.mjs" $ARGUMENTS
Parse the JSON output. The result contains:
valid: boolean — whether the trace passed validationformat: "otel" | "claude-code" | "unknown"errors: array of critical failure messageswarnings: array of non-critical issuesmetadata: object with format-specific fields
If valid is true, display the validation results:
For OTEL:
Trace Validation: PASSED
Format: OTEL
Trace ID: <metadata.traceId>
Service Name: <metadata.serviceName>
Span Count: <metadata.spanCount>
System Prompt: <metadata.systemPromptDetected ? "FOUND (via " + metadata.systemPromptMethod + ")" : "NOT FOUND">
For Claude Code:
Trace Validation: PASSED
Format: Claude Code
Session ID: <metadata.sessionId>
Event Types: <metadata.eventTypes joined with ", ">
System Prompt: <metadata.systemPromptDetected ? "FOUND (via " + metadata.systemPromptMethod + ")" : "NOT FOUND">
If valid is false, check whether the errors are recoverable (eligible for flexible validation):
Recoverable errors (any of these present → offer fallback):
- "No system prompt found"
- "Unknown trace format"
- "Invalid Claude Code event structure"
Unrecoverable errors (if ANY of these are present → do NOT offer fallback):
- "File not found"
- "File is not valid JSON or JSONL"
- "Empty Claude Code trace"
- "Missing resourceSpans"
- "No spans found in trace"
If any unrecoverable error is present, display the errors and stop:
Trace Validation: FAILED
Errors:
- <each error on its own line>
If at least one error is recoverable, offer flexible validation via AskUserQuestion:
- question:
Deterministic validation failed: ───────────────────────────────────────────────────────── <each error on its own line, prefixed with •> ───────────────────────────────────────────────────────── Would you like to try flexible (AI-assisted) validation? This uses an LLM to assess whether the trace is still structurally usable despite not passing strict format checks. - header: "Flexible Validation"
- options:
{ "label": "Try flexible validation", "description": "Use AI to assess trace structure" }{ "label": "Stop", "description": "Accept the validation failure" }
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.
- 2d ago First seen · 131 lines · 14 tokens per session scan A e54d213a47c6
validate is a command published in the GitHub repository Not-Diamond/self-care (28 stars, last pushed 4mo ago), licensed MIT. It adds 14 tokens to every session and 946 once invoked, about $0.0001 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 commands, from other repositories
coder-eval-implement-plan
Implement an approved codereval plan phase by phase with risk-scaled per-phase review, then a final code review.
VibeGuard: Stats
View hooks trigger statistics - interception/warning/release times and reason analysis.
closeout
Close out a UNITARES workspace before final response.
checkin
Manual UNITARES governance check-in for the current turn.
re-review
Round 2+ review — confirm revisions landed, catch regressions, do not re-litigate.
project
Follow the project management workflow defined in .agents/skills/project/SKILL.md.