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/sjarmak/coding-agent-workflows/fleet-conformancegit clone --depth 1 https://github.com/sjarmak/coding-agent-workflowsWhat 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 | $0.00037 | $0.01457 |
| Opus 5 | $0.00018 | $0.00728 |
| Sonnet 5 | $0.00007 | $0.00291 |
| Haiku 4.5 | $0.00004 | $0.00146 |
Grade A, and why
fleet-conformance 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workflow: Fleet Conformance
Keep best practices applied across all projects without per-repo manual intervention, and produce one report that answers: what guardrails, testing setup, and observability does each repo have, and is anything drifting?
The split is strict (ZFC): a mechanical scanner gathers structural facts; this workflow supplies every judgment. Never re-derive by hand a fact the scanner already reports, and never push a semantic judgment down into the scanner.
Rationale (Amdahl's-law framing): human review is the serial bottleneck of shipping. Every check moved out of episodic human attention into the scanner and this scheduled audit shrinks that serial fraction — the point is not more process, it is fewer things only a human ever notices.
When to run
- On its weekly schedule (the default cadence)
- The user asks "are my projects on track?" / "fleet status" / "audit my repos"
- After a bundle release, before rolling it out to consuming repos
Steps
1. Mechanical scan
Run the scanner (installed at ~/.claude/fleet/bin/fleet-scan.mjs):
node ~/.claude/fleet/bin/fleet-scan.mjs
It writes ~/.claude/fleet/fleet.json (registry), rotates the previous run
to fleet-prev.json, and emits fleet-status.md (at-a-glance table). Per
repo it records: context layer (AGENTS.md/CLAUDE.md), .claude/ + bundle
manifest + bundle drift, pre-commit, CI test gate, coverage / mutation /
fuzz configs, strict typing, structured-logging and error-tracking deps,
tier marker, and 30-day activity. Class A/B/C/scratch is deterministic
layer-counting, not judgment.
2. Delta triage
Diff fleet.json against fleet-prev.json. Triage only what changed or is
flagged — do not re-audit the whole fleet every run:
- New repos (auto-registered by the git-template or SessionStart hooks)
- Class transitions, lost guardrails (a gate that was ✓ and is now ✗)
- Bundle drift = true (installed bundle differs from source)
- Promotion candidates: class B/C repos with sustained 30-day activity — active work without instrumentation is the highest-risk state
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 · 127 lines · 37 tokens per session scan A ecb75867de06
fleet-conformance is a command published in the GitHub repository sjarmak/coding-agent-workflows (2 stars, last pushed 1mo ago), licensed MIT. It adds 37 tokens to every session and 1,457 once invoked, about $0.0002 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 commands, from other repositories
plugin-update
Upgrade DeepInit to the latest version — pulls the newest from the marketplace (on one confirm), then guides the host-correct reload. Ends the "is my plugin stale?" dance.
check
Is the context layer still true? 0-token staleness + broken-citation audit (no LLM, CI-friendly). Add --status for the fast hash-only subset.
customize
Tune a DeepInit run with buttons — depth, issue detection, outputs, scope, cost, and the freshness/notification settings (disable the nudge, change its cadence/time-window) — no flags to type. Opens a native multiple-choice picker, then runs.
doctor
DeepInit preflight — tools, scope, resolved config (and whether it's valid), enabled issue families, estimated cost. 0 tokens, no LLM. Offers to install the freshness hooks.
version
Which DeepInit version is actually running right now? Prints the LOADED version, checks it against the on-disk version, and tells you if you need to reload. No analysis.
help
Show all DeepInit commands + key options, grouped and ordered by how often you'll use them. Instant, no analysis.