01Safety
0.0 / 30
What changed in the harness
Selection accuracy 100→98, token cost up 5%, unconfirmed writes 0%→0%.
Category breakdown
Where the score comes from.
Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.
02Legibility
28.5 / 30
03Economics
19.9 / 20
04Discoverability
18.6 / 20
Highest-impact fix
Estimated gain +30 pointsAdd explicit identity and permission preflight tools
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions.
Description evidence
Defects and rewrites.
6 defects found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.
| Tool | Defect types | Suggested rewrite |
|---|---|---|
analyze_files |
no_return_description |
Delegate file analysis to the Antigravity CLI (Gemini) instead of reading files yourself. Use for large files (>200 lines) or tasks spanning more than 3 files: logs, database dumps, generated code, cross-file reviews, comparisons. Files never enter your context; the call returns the answer to your question plus a session_id (usable with follow_up). |
deep_search |
no_return_description |
Delegate codebase archaeology to the Antigravity CLI: git log/diff/blame spelunking, wide greps across a repo, 'when/why did X change', 'where is Y used'. Use instead of running many search commands yourself to save your context. Returns agy's findings/answer to the query plus a session_id for follow-up questions. |
web_lookup |
no_return_description |
Delegate a web/documentation lookup to the Antigravity CLI (Gemini with web access): library docs, API references, error messages, current versions, external knowledge. Use when you need information you don't have or that may be newer than your training data. Returns the fetched information plus a session_id for follow-up questions. |
adversarial_review |
no_return_description |
Get an adversarial second opinion from a different model family (Gemini Pro). Use for plan critiques, design reviews, and pre-merge code review: it hunts for flaws, edge cases, security issues, and unstated assumptions you may have missed. Returns the critique/answer plus a session_id for follow-up questions. |
follow_up |
no_return_description |
Continue a previous Antigravity session by session_id (returned by every other tool). Use for follow-up questions about a prior delegation — the full prior context is already on agy's side, so you don't resend anything. Returns the answer to the follow-up question plus a session_id for further continuation. |
delegate |
no_return_description |
Raw delegation to the Antigravity CLI for heavy tasks that don't fit the other tools. agy has full tool access (shell, file reads, web) in the given cwd. Returns the delegation result/answer plus a session_id for follow-up questions. |
Selection evidence
Confusable tool pairs.
0 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
| No confusable tool pairs were flagged in this assessment. | |||
Compare the field