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/bar181/bar-observatory/bar-reportgit clone --depth 1 https://github.com/bar181/bar-observatoryWrote 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/bar181/bar-observatory/bar-report)<a href="https://agentmods.dev/commands/bar181/bar-observatory/bar-report"><img src="https://agentmods.dev/badge/commands/bar181/bar-observatory/bar-report.svg" alt="Measured on agentmods" 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 | $0.00019 | $0.00242 |
| Opus 5 | $0.00010 | $0.00121 |
| Sonnet 5 | $0.00004 | $0.00048 |
| Haiku 4.5 | $0.00002 | $0.00024 |
Grade A, and why
bar-report 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 6d 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.
What it actually says
$ARGUMENTS
Generate BAR Observatory's deterministic engineering report — no LLM in the render path, same database in, byte-identical output out.
- If
$ARGUMENTSnames a database path, use it. Otherwise default to.bar/ambient.sqliterelative to the current project (the pathbar init --dir .creates); if that file doesn't exist, tell the user to runbar init --dir .thenbar ingest .bar/ambient.sqlite <session>.jsonlfirst, and stop. - Run:
bar report <db> --out . - Report the three output paths it wrote (
*.report.json,*.report.html,*.report.md) and tell the user to open the.htmlfile for the human view. - If the command fails, run
bar doctor <db>and surface what it reports before guessing at the cause — a missing or corrupt store is the most common reasonbar reportfails.
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.
- 6d ago First seen · 19 lines · 19 tokens per session scan A 5509d68be392
bar-report is a command published in the GitHub repository bar181/bar-observatory (22 stars, last pushed 9d ago), licensed MIT. It adds 19 tokens to every session and 242 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
sync
Sync the audit manifest with Azure DevOps work items — set the connector up for the first time (connect: verify the transport, report which auth path is in effect, prove access read-only, detect the board's process), push manifest bugs/tasks/phases to ADO (board states, sprint stamp, Remaining Work, comments), pull…
init
Multi-agent codebase audit that GENERATES the audit manifest (phases/tasks) at manifestPath. Interviews you for scope/goals, fans out parallel read-only explorers, synthesizes findings, then presents the proposed phases for approval BEFORE writing — approve to materialize, or park them as proposals for later…
task
Add a tracked task to the audit manifest — every answer is a flag, and the dialogue only covers what the caller did not pass — move one between phases, or cancel work that will not be done. add allocates the id, initializes all orchestrator fields, updates fileIndex, and revalidates; move renumbers a task into another…
panel
Audit pipeline: open / stop / check a local control-panel UI to visually manage .claude/audit.config.json and the manifest's composition levers (reviewSkill, per-task skills/models, buildCommands) — with live validation and discovery of the skills & agents available in this repo + globally. Ephemeral, on-demand; a…
doctor
Audit pipeline: diagnose the setup before it bites — interpreter the hooks will use, git root, config, manifest + shard integrity, which plan-gate tier is active, submodule conflicts, build runners, whether the skills the plan names would resolve from a clone or only here, whether hooks have ever fired and which copy…
status
Audit pipeline: print manifest status — phases, tasks, bugs, the ready-now list and what each pending task is waiting on; or, with --gate, turn that same state into a CI pass/fail verdict over conditions you pick with --fail-on. Read-only, no locks, no mutations.