Loading ToolBench

Reading the current corpus…

instagram-mcp MCP server · D · ToolBench

instagram-mcp

github.com/dr-ashraf-s-MCP-LABS/instagram-mcp

MCP server to control Instagram from Claude AI - post photos, reels, carousels, reply to comments, send DMs, and get analytics

2026-07-28+STDIO5 tools

58/100

Definition Quality74

Protocol Readiness50

Supportability36

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

Server has 5 tools with clear verb-noun naming (get_*, list_*, update_*). Descriptions are present and moderately detailed (100-250 chars), explaining WHAT and WHEN. Input schemas are visible with types and constraints (enums, min/max). However, output schemas are NOT documented, LLMs cannot predict return structure. Parameter descriptions are adequate but lack dependency hints. The update_profile tool correctly flags WRITE operations. Error handling is basic (ValueError with context) but lacks recovery guidance or categorization. No tool annotations (readOnlyHint/destructiveHint) despite clear risk levels.

Tools (5)

get_account_inforead onlyauthsource verified75/100

Fetch basic info about the connected Instagram Business account. Use this first to verify the token and IG user ID are valid. Returns username, follower count, media count, and account type.

get_account_insightsread onlyauthsource verified72/100

Get account-level insights (reach, impressions, profile views, follower growth). Use 'day' for last 24h, 'week' for 7 days, 'days_28' for monthly trends. Useful for content strategy decisions.

get_media_insightsread onlyauthsource verified70/100

Get performance metrics for a single post/reel. Returns reach, impressions, saved, video_views (if video), and engagement. Different media types return different metrics — reels have plays, photos don't.

list_recent_mediaread onlyauthsource verified78/100

List recent posts/reels from the account. Returns id, caption, media_type, permalink, timestamp, like_count, comments_count. Use the returned IDs with get_media_insights or get_comments.

update_profilewriteauthsource verified77/100

Update your Instagram Business profile bio and/or website URL. THIS IS A WRITE OPERATION. At least one of biography or website must be provided. Changes are live immediately on your public profile. Required permission: instagram_business_manage_profile (Add this in your Meta app → Use cases → Permissions and features if missing.)

Issues (5)

HIGHtool ↗

Output schemas not documented. LLMs cannot predict return structure (fields, types, nesting). Responses are inferred from code comments only.

get_account_infolist_recent_mediaget_media_insightsget_account_insightsupdate_profile

MEDIUMtool-annotations ↗

No tool annotations (readOnlyHint, destructiveHint, idempotentHint). update_profile is marked WRITE in metadata but tool definition lacks destructiveHint annotation.

update_profile

MEDIUMrecovery-guide ↗

Error handling returns raw ValueError with context string. No recovery guidance, categorization (retryable vs user-fixable), or actionable next steps for LLM.

get_account_infolist_recent_mediaget_media_insightsget_account_insightsupdate_profile

MEDIUMpaginated-result ↗

list_recent_media accepts limit (1-50) but no pagination support (offset, cursor, next_token). Large result sets not handled; no total_count returned.

Recommendations

Add explicit output schema documentation for each tool. Example: 'Returns dict with keys: username (str), followers_count (int), media_count (int), profile_picture_url (str).'
Add tool annotations to tool definitions: @mcp.tool(annotations={'readOnlyHint': True}) for read tools; @mcp.tool(annotations={'destructiveHint': True}) for update_profile.
Enhance error handling: return structured error objects with 'error_code', 'message', 'recovery_hint', and 'retryable' fields. Example: {'error_code': 'INVALID_TOKEN', 'message': '...', 'recovery_hint': 'Refresh META_ACCESS_TOKEN in .env', 'retryable': False}.
Add pagination to list_recent_media: accept 'after' cursor parameter and return 'next_cursor' in response. Document: 'Returns up to 50 items; use next_cursor to fetch more.'
Document update_profile constraint explicitly: 'At least one of biography or website must be non-null. If both are null, returns error: "At least one field required."'

Score history

Overall score trend

First recorded score · v2 rubric

58/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	D	58	2026-07-28+	v2