grants-gov-monitor — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited grants-gov-monitor (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
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.
You build a pipeline of federal grant opportunities matched to your organization's eligibility, capability, and capacity. The 2026 landscape: 2,200+ federal assistance programs (per SAM.gov assistance listings, which replaced CFDA). GrantMetric reports ~900+ active opportunities across NIH/NSF/DOD/USDA/EPA/etc. The discipline isn't finding grants — it's filtering them.
============================================================ === PRE-FLIGHT === ============================================================
Recovery:
============================================================ === PHASE 1: OPPORTUNITY INGESTION === ============================================================
Pull from multiple sources daily:
Grants.gov API (open, no key required):
GET https://api.grants.gov/v1/api/search2
?keyword={query}
&oppStatuses=forecasted|posted
&eligibilities={code_list}
&fundingCategories={cat_list}
&agencies={agency_list}SAM.gov Assistance Listings: 2,200+ programs with CFDA numbers (still used as program identifiers post-CFDA sunset). Use SAM.gov API.
Agency-specific portals:
Persist all opportunities to a local DB with schema:
{
"opp_id": "...",
"source": "GRANTS_GOV|SAM_AL|NIH|NSF|DOE|...",
"title": "...",
"agency": "...",
"cfda_number": "93.279",
"summary": "...",
"eligibility_codes": ["...", "..."],
"category": ["...", "..."],
"funding_total_estimate": 5000000,
"award_floor": 100000,
"award_ceiling": 500000,
"expected_award_count": 10,
"post_date": "2026-04-01",
"loi_due": "2026-06-15",
"full_app_due": "2026-08-15",
"url": "...",
"match_required_pct": 25,
"indirect_cap_pct": null,
"geographic_restriction": null,
"performance_period_months": 36
}VALIDATION: Daily ingestion captures > 95% of newly-posted opportunities. Diff vs previous day produces a clean "new today" list.
============================================================ === PHASE 2: ELIGIBILITY FILTER === ============================================================
Filter the universe to opportunities the user can actually apply for:
| Eligibility Code | Org Types |
|---|---|
| 00 | Unrestricted |
| 01 | County governments |
| 02 | City or township |
| 04 | Special district |
| 05 | Independent school district |
| 06 | Public/State controlled IHE |
| 07 | Native American tribal organizations (federally recognized) |
| 08 | Public housing authorities |
| 11 | Native American tribal organizations (non-federally recognized) |
| 12 | Nonprofits with 501(c)(3) status |
| 13 | Nonprofits without 501(c)(3) status |
| 20 | Private institutions of higher education |
| 21 | Individuals |
| 22 | For-profit organizations (non-small businesses) |
| 23 | Small businesses |
| 25 | Others |
| 99 | Unrestricted |
Map user's org type → eligibility codes → opportunities. Hide non-eligible. Surface "near eligible" (e.g., need fiscal sponsor) separately.
VALIDATION: Filter precision ≥ 95% (no false-positive "eligible" opportunities).
============================================================ === PHASE 3: CAPABILITY MATCH SCORING === ============================================================
Beyond eligibility, score each opportunity on capability fit:
| Dimension | Score 1-5 | Notes |
|---|---|---|
| Subject-matter fit | 1 = adjacent, 5 = on-mission | Cosine similarity of opp summary vs org capability statement |
| Prior award credibility | 1 = no past, 5 = active funded researcher in the program | Pull from NIH RePORTER / USAspending |
| Award size fit | 1 = too small to bother / too big to credibly win, 5 = right size for org capacity | $25k-$1M sweet spot for small nonprofits |
| Deadline reality | 1 = < 30 days (rushed), 5 = > 90 days | Quality proposals need 6-12 weeks |
| Match feasibility | 1 = can't meet match, 5 = no match required | Tag opps with > 25% match if user lacks capacity |
| Competition density | 1 = R01 (10% success rate), 5 = niche small business (40%+) | Pull historical award rate per program |
Composite Pwin = weighted average. Pursue threshold: ≥ 3.5.
VALIDATION: Each opp has Pwin score with rationale per dimension.
============================================================ === PHASE 4: DEADLINE CALENDAR === ============================================================
Generate calendar with critical dates per opportunity:
Output grants_calendar.ics. Reminder: 12 weeks / 8 weeks / 4 weeks / 2 weeks / 1 week before each deadline.
VALIDATION: Calendar imports cleanly. Reminders fire at expected intervals.
============================================================ === PHASE 5: SAM.GOV + UEI READINESS === ============================================================
Before applying, verify:
Generate sam_uei_health_check.md flagging any gaps.
VALIDATION: SAM.gov registration verified within 30 days. Expired registration blocks federal awards.
============================================================ === PHASE 6: PROPOSAL READINESS SCAFFOLD === ============================================================
For each "pursue" opportunity, generate the proposal-prep package:
Output proposal_kit/{opp_id}/ directory.
VALIDATION: Kit has all standard sections; agency-specific templates populated.
============================================================ === PHASE 7: WEEKLY DIGEST === ============================================================
Schedule weekly Monday digest:
# Grants Pipeline — Week of {date}
## New opportunities this week ({N})
{Filtered list with Pwin ≥ 2 — surface broader pool, even if below pursue threshold}
## Pursue queue ({M})
{Opps with Pwin ≥ 3.5 + LOI/app deadline within 12 weeks}
## Deadlines this month
{Calendar excerpt}
## Closed / awarded
{Recent awardees in your topic area (intelligence)}
## Recommended actions
{1-3 specific next steps — e.g., "Email NSF program officer X re: opp Y", "Begin biosketch updates for {personnel}", "Confirm NICRA covers indirect on opp Z"}VALIDATION: Digest fits one screen; actions are concrete.
============================================================ === SELF-REVIEW === ============================================================
Common gap: SAM.gov registration expiring mid-application cycle → reject. Verify annually with explicit reminder.
============================================================ === LEARNINGS CAPTURE === ============================================================
~/.claude/skills/grants-gov-monitor/LEARNINGS.md.
============================================================ === STRICT RULES === ============================================================
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.