Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Configure Fireflies.ai GraphQL API authentication and verify connectivity. Use when setting up a new Fireflies.ai integration, configuring API keys, or initializing the GraphQL client for transcript access. Trigger with phrases like "install fireflies", "setup fireflies", "fireflies auth", "configure fireflies API key".
.claude/skills/jeremylongshore-fireflies-install-auth/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -10% | 0% |
Create an auditable identity boundary before any transcript query or mutation. A successful request proves only that the key is accepted; it does not grant permission to enumerate or export meeting content.
Fireflies uses POST requests to https://api.fireflies.ai/graphql with Content-Type application/json and Authorization: Bearer <API key>. Obtain the key from the Fireflies Integrations page, keep it server-side, and use the smallest identity query needed to confirm the principal and team context.
For authenticated operations, inject FIREFLIES_API_KEY from an approved secret manager and send it only as Authorization: Bearer REDACTED_KEY to https://api.fireflies.ai/graphql. Never print, commit, place in a URL, forward to a browser, or include the key in evidence. Webhook signing secrets are separate credentials and must not be reused as API keys.
Use Read, Glob, and Grep to inspect code, configuration, tests, and evidence. Use Write/Edit only for approved implementation or documentation changes. Do not query Fireflies, retrieve meeting content, create an AskFred thread, upload media, change account state, replay an event, or deploy merely because this skill was invoked.
Require approval before creating, rotating, revoking, or broadening a production key, changing its owning account, or querying any real meeting record.
Return the exact operation or event surface, environment, authorization class, selected field groups, validation results, content-free metrics, decisions, and a concise pass/fail receipt. Keep secrets and meeting-derived content out of general output.
Before reporting success, rerun the smallest relevant deterministic check, compare actual state with the requested outcome and current contract, verify no secret or meeting-derived content entered logs or artifacts, and record unresolved uncertainty explicitly.
Read official Fireflies.ai evidence before relying on a field, filter, event, permission, plan limit, mutation, or processing state.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 739,133 | 18,084 | -98% | 1 | 1 | 0% | 2,372 | 2,953 | +24% | 0 | 0 | — |
case-02 | fail→pass | 10,986 | 5,480 | -50% | 1 | 1 | 0% | 1,692 | 2,311 | +37% | 0 | 0 | — |
case-03 | pass→pass | 10,087 | 5,740 | -43% | 1 | 1 | 0% | 1,898 | 2,302 | +21% | 0 | 0 | — |
case-04 | pass→pass | 11,452 | 7,179 | -37% | 1 | 1 | 0% | 2,303 | 2,582 | +12% | 0 | 0 | — |
case-05 | fail→pass | 11,043 | 10,694 | -3% | 1 | 1 | 0% | 2,298 | 3,530 | +54% | 0 | 0 | — |
case-06 | fail→pass | 13,698 | 6,557 | -52% | 1 | 1 | 0% | 2,245 | 2,247 | +0% | 0 | 0 | — |
case-07 | pass→pass | 6,903 | 3,031 | -56% | 1 | 1 | 0% | 1,033 | 1,665 | +61% | 0 | 0 | — |
case-08 | fail→pass | 13,583 | 3,853 | -72% | 1 | 1 | 0% | 2,043 | 1,837 | -10% | 0 | 0 | — |
case-09 | pass→pass | 11,287 | 7,759 | -31% | 1 | 1 | 0% | 1,929 | 2,595 | +35% | 0 | 0 | — |
case-10 | pass→pass | 15,014 | 14,137 | -6% | 1 | 1 | 0% | 2,633 | 3,645 | +38% | 0 | 0 | — |
case-11 | pass→pass | 4,815 | 2,574 | -47% | 1 | 1 | 0% | 788 | 1,550 | +97% | 0 | 0 | — |
case-12 | pass→pass | 1,918 | 1,557 | -19% | 1 | 1 | 0% | 291 | 1,431 | +392% | 0 | 0 | — |
case-13 | pass→pass | 6,649 | 1,815 | -73% | 1 | 1 | 0% | 1,084 | 1,449 | +34% | 0 | 0 | — |
case-14 | pass→pass | 6,770 | 1,815 | -73% | 1 | 1 | 0% | 1,251 | 1,465 | +17% | 0 | 0 | — |
case-15 | pass→pass | 10,331 | 3,037 | -71% | 1 | 1 | 0% | 1,930 | 1,693 | -12% | 0 | 0 | — |
case-16 | pass→pass | 11,620 | 3,030 | -74% | 1 | 1 | 0% | 1,952 | 1,628 | -17% | 0 | 0 | — |
case-17 | pass→pass | 6,812 | 2,564 | -62% | 1 | 1 | 0% | 1,125 | 1,517 | +35% | 0 | 0 | — |
case-18 | pass→fail | 6,769 | 6,407 | -5% | 1 | 1 | 0% | 992 | 2,218 | +124% | 0 | 0 | — |
case-19 | pass→pass | 15,898 | 15,956 | +0% | 1 | 1 | 0% | 2,678 | 4,067 | +52% | 0 | 0 | — |
case-20 | fail→fail | 9,051 | 16,254 | +80% | 1 | 1 | 0% | 1,716 | 4,599 | +168% | 0 | 0 | — |
case-21 | fail→fail | 13,892 | 18,979 | +37% | 1 | 1 | 0% | 2,546 | 4,824 | +89% | 0 | 0 | — |
case-22 | fail→fail | 12,964 | 19,273 | +49% | 1 | 1 | 0% | 2,319 | 5,019 | +116% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +18 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.