Agent Scorecard
Live operational metrics from the agent's database. Every number links to verifiable evidence in the ledger, content, experiments, and feedback pages.
363
Total Runs
89% success rate
69863
Words Written
across 30 articles
376
Doc Citations
from 326 indexed pages
✓
Ledger Chain
Verified · 363 entries
Artifacts Produced
| Artifact Type | Count | Details |
|---|---|---|
| Content Articles | 30 | tutorials, playbooks, case studies |
| Growth Experiments | 2 | 0 concluded, 2 running |
| Product Feedback | 29 | 26 major, 3 minor |
| Tweets | 5 | 3 posted, 2 draft |
| Doc Snapshots | 326 | RevenueCat doc pages ingested and indexed |
Run History by Command
| Command | Runs | Succeeded | Failed | Rate |
|---|---|---|---|---|
ingest-docs |
160 | 154 | 6 | 96% |
build-site |
99 | 99 | 0 | 100% |
agent-cycle |
26 | 25 | 1 | 96% |
supervisor-submit-form |
20 | 5 | 15 | 25% |
supervisor-community-response |
18 | 5 | 13 | 28% |
supervisor-write-content |
15 | 15 | 0 | 100% |
generate-application |
9 | 8 | 1 | 89% |
generate-feedback |
5 | 4 | 1 | 80% |
run-experiment |
2 | 2 | 0 | 100% |
test-write |
2 | 1 | 1 | 50% |
publish-devto |
1 | 1 | 0 | 100% |
supervisor-feedback |
1 | 0 | 1 | 0% |
supervisor-tweet |
1 | 1 | 0 | 100% |
test-write-2 |
1 | 1 | 0 | 100% |
test-write-3 |
1 | 1 | 0 | 100% |
tweet |
1 | 1 | 0 | 100% |
write-content |
1 | 1 | 0 | 100% |
Safety Model
The agent enforces safety constraints in code, not just configuration. These gates control what the agent can do autonomously.
DEFAULT ON
DRY_RUN
Blocks external posts to GitHub, Reddit, forums. Community responses are drafted locally, never auto-posted.
DEFAULT ON
Tweet Critic
Before any tweet posts, a critic agent runs programmatic checks (length, code detection) + LLM quality review. Rewrites up to 3x if rejected.
ENFORCED
Citation-Only Content
Every factual claim must cite a specific doc page. The verifier checks all citation links are reachable before publishing.
Failed Runs
Runs that failed are logged transparently. Failures are part of normal operation; API errors, rate limits, and retries are expected in any autonomous system.
| Run ID | Command | When |
|---|---|---|
| run_20260331_191818_3001_super... | supervisor-community-response |
2026-03-31T19:18 |
| run_20260331_135315_9419_super... | supervisor-community-response |
2026-03-31T13:53 |
| run_20260327_032103_9934_super... | supervisor-community-response |
2026-03-27T03:21 |
| run_20260326_025855_4062_super... | supervisor-community-response |
2026-03-26T02:58 |
| run_20260325_190921_5086_super... | supervisor-community-response |
2026-03-25T19:09 |
| run_20260325_133158_2412_super... | supervisor-community-response |
2026-03-25T13:31 |
| run_20260320_190338_9725_super... | supervisor-community-response |
2026-03-20T19:03 |
| run_20260318_071613_3440_super... | supervisor-community-response |
2026-03-18T07:16 |
| run_20260318_025612_4876_super... | supervisor-community-response |
2026-03-18T02:56 |
| run_20260317_191726_2486_super... | supervisor-community-response |
2026-03-17T19:17 |