01Safety
0.0 / 30
What changed in the harness
Selection accuracy 100→100, token cost up 3%, 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
28.8 / 30
03Economics
19.6 / 20
04Discoverability
14.9 / 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.
1 defect 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 |
|---|---|---|
convert_tiktok_unicode_font |
no_return_description |
Convert plain text into a Unicode font style for TikTok bios, captions, and on-screen text. Returns the converted text in the requested style. Supports bold, italic, boldItalic, cursive, monospace, doubleStruck, and fraktur. |
Selection evidence
Confusable tool pairs.
4 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
calculate_tiktok_rpm |
calculate_tiktok_engagement_rate |
medium | Both are 'calculate_tiktok_*' analytics tools that take 'views' as an input, so a task like 'calculate my TikTok performance metrics from these views' or 'what will 500k views earn/convert to' is ambiguous; revenue estimation vs engagement percentage both plausibly match vague metric-calculation wording. |
get_tiktok_hashtags |
get_tiktok_trends_by_country |
medium | A task like 'get the top/trending TikTok hashtags' matches both: hashtags returns top 15 hashtags by niche, while trends_by_country also returns trending hashtags (plus music/videos). Users rarely specify niche vs country, so 'what hashtags are hot right now' can route to either tool. |
get_tiktok_hashtags |
get_tiktok_keyword_research |
medium | Users commonly conflate 'hashtags' and 'keywords' for content planning (e.g., 'get TikTok keywords/tags for my fitness niche'), and both serve content-strategy: hashtags takes a niche while keyword_research takes a seed keyword, so a vague 'research tags/keywords for my content' task is genuinely ambiguous. |
download_tiktok_video |
extract_tiktok_audio_mp3 |
low | Both accept a TikTok video URL for a 'save/grab this TikTok' style task, so vague wording like 'download this TikTok' or 'get the sound off this video' can route to the wrong member, though the descriptions explicitly separate video download vs audio extraction, keeping confusion only mildly plausible. |
Compare the field