monthly-burn — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited monthly-burn (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
The thesis: a founder cannot manage what they cannot see. Cost visibility per source per month is the table that decides which experiments live and which die. The skill reads MCP-exposed sources where possible; everything else surfaces as Cooper-manual-entry so the gap is visible.
bash skills/monthly-burn/run.sh # current month
bash skills/monthly-burn/run.sh --month 2026-04 # specific month
bash skills/monthly-burn/run.sh --dry-runFor each cost source:
data/finance/manual/anthropic-<YYYY-MM>.txt if it exists (Cooper paste). Otherwise Cooper-manual-entry.data/voice/elevenlabs_usage.json if present + token-cost calc.verify-license deploys, read CF metrics.data/finance/manual/stripe-<YYYY-MM>.txt.Cooper-manual-entry.data/finance/domain-registrations.json (annual entries; pro-rated to month).Sum to a monthly total. Flag any Cooper-manual-entry rows.
# Monthly burn — <YYYY-MM>
## Sources
| Source | This month | Source-of-truth | Status |
|---|---|---|---|
| Anthropic | $X | data/finance/manual/anthropic-... | automated/manual |
| ElevenLabs | $Y | data/voice/elevenlabs_usage.json | automated |
| Cloudflare | $0 | n/a (worker not deployed) | placeholder |
| Stripe fees | $Z | Stripe MCP / manual | automated/manual |
| GitHub Actions | $W | GitHub MCP / manual | automated/manual |
| Domain registrations | $V | data/finance/domain-registrations.json | automated |
| **Total burn** | $T | | |
## Manual-entry gaps
- <list of sources that fell back to Cooper-manual-entry>
## Trend
- Last month total: $T_prev
- Delta: ±$DSTATUS=ok month=<YYYY-MM> total=<dollars> manual_gaps=<N> / STATUS=skip reason=<...> / STATUS=error reason=<...>
Each cost source is queried via MCP tools or file reads. No MCP server is being authored here — the aggregator skill is just bash + jq calling the existing MCP-server-exposed tools or reading files. Skill, not Python.
If the latest data/finance/burn/<YYYY-MM>.md is older than 35 days (one month + buffer), the heartbeat-silence pattern catches it.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.