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.
git clone --depth 1 https://github.com/Enovatr-Labs/SpecRouteWrote 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/commands/enovatr-labs/specroute/audit)<a href="https://agentmods.dev/commands/enovatr-labs/specroute/audit"><img src="https://agentmods.dev/badge/commands/enovatr-labs/specroute/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/commands/enovatr-labs/specroute/audit"><img src="https://agentmods.dev/badge/commands/enovatr-labs/specroute/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 | $0.00024 | $0.03398 |
| Opus 5 | $0.00012 | $0.01699 |
| Sonnet 5 | $0.00005 | $0.00680 |
| Haiku 4.5 | $0.00002 | $0.00340 |
Grade B, and why
audit scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletemediumDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf "$work" Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
raw = subprocess.check_output( How it starts
The opening of the file, as written. The whole thing — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run a comprehensive audit. This is the gate before any push or PR.
1. Sanitization
Run /sanitize first. If it fails, stop and report the failures - do not continue with later checks until sanitization is clean.
2. Frontmatter validation
Walk every agent / skill / command file and validate frontmatter:
Contracts are vendor-scoped since the mid-2026 convergence - see wiki/Frontmatter-Contracts.md and the frontmatter-lint skill. Do not hold non-Claude vendors to Claude's contract.
echo "── frontmatter check (agents: Claude contract) ──"
# Claude requires name + description; model and color are optional.
for f in .claude/agents/*.md agents/agent-template.md agents/examples/*.md runtimes/.claude/agents/*.md; do
[ -f "$f" ] || continue
case "$f" in *README.md) continue ;; esac
for k in name description; do
grep -q "^$k:" "$f" || echo " ERROR $f: missing $k"
done
done
echo
echo "── frontmatter check (agents: Codex contract) ──"
# Codex agents are TOML, not Markdown - keys are `name = "..."`, not `name:`.
for f in runtimes/.codex/agents/*.toml; do
[ -f "$f" ] || continue
for k in name description developer_instructions; do
grep -qE "^$k *=" "$f" || echo " ERROR $f: missing $k"
done
done
echo
echo "── frontmatter check (agents: other vendors) ──"
# SpecRoute's portable publication baseline is name + description. Some vendor
# parsers accept less; the audit intentionally holds shipped templates to the
# stronger portable baseline. model/color remain Claude-specific.
# Devin Local subagent profiles use per-profile `AGENT.md` directories under
# `runtimes/.devin/agents/`. Cascade compatibility paths are not a second runtime.
for f in runtimes/.gemini/agents/*.md runtimes/.kiro/agents/*.md runtimes/.cursor/agents/*.md runtimes/.devin/agents/*/AGENT.md; do
[ -f "$f" ] || continue
case "$f" in *README.md) continue ;; esac
for k in name description; do
grep -q "^$k:" "$f" || echo " ERROR $f: missing $k"
done
done
echo
echo "── frontmatter check (skills) ──"
# Contract lives in wiki/Frontmatter-Contracts.md. The Agent Skills open standard
# uses name + description as the portable baseline. argument-hint /
# user-invocable / allowed-tools are optional Claude Code extensions, so they
# are warnings only in Claude-shaped templates.
for f in .claude/skills/*/SKILL.md .agents/skills/*/SKILL.md \
skills/skill-template/SKILL.md skills/examples/*/SKILL.md \
runtimes/.claude/skills/*/SKILL.md runtimes/.codex/skills/*/SKILL.md \
runtimes/.gemini/skills/*/SKILL.md runtimes/.kiro/skills/*/SKILL.md \
runtimes/.cursor/skills/*/SKILL.md runtimes/.devin/skills/*/SKILL.md; do
[ -f "$f" ] || continue
for k in name description; do
grep -q "^$k:" "$f" || echo " ERROR $f: missing $k"
done
case "$f" in
.claude/*|skills/*|runtimes/.claude/*)
for k in argument-hint user-invocable allowed-tools; do
grep -q "^$k:" "$f" || echo " WARN $f: missing $k (optional Claude extension)"
done ;;
esac
done
echo
echo "── frontmatter check (Claude commands) ──"
for f in commands/command-template.claude.md commands/examples/*.claude.md runtimes/.claude/commands/*.md .claude/commands/*.md; do
[ -f "$f" ] || continue
case "$f" in *README.md) continue ;; esac
grep -q "^description:" "$f" || echo " ERROR $f: missing description"
done
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 · 249 lines · 24 tokens per session scan B 62d8fb793a4b
audit is a command published in the GitHub repository Enovatr-Labs/SpecRoute (3 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 24 tokens to every session and 3,398 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 2 findings (recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
sdd-continue
Continue the next SDD phase in the dependency chain.
sdd-apply
Implement SDD tasks — writes code following specs and design.
sdd-explore
Explore and investigate an idea or feature — reads codebase and compares approaches.
gentle-sdd-ff
Fast-forward all SDD planning phases — proposal through tasks.
review-branch
Review the current branch's diff against base by dispatching atomic-reviewer. No orchestration loop, no spec required — pre-flight before /commit pr or /commit merge.
address-pr-feedback
Systematically address PR review comments. Fetches all threads, categorizes by status (acknowledged, silently fixed, unaddressed), and guides user through posting replies and implementing fixes with explicit approval.