Scores 7.2/10 overall. with execution at 7.8 and access readiness at 6.4. Payment: Slack app distribution via marketplace; billing API limited. Governance: Enterprise Grid has strong RBAC and audit; free/pro tiers limited. Web accessibility: Complex web UI; bot/app APIs are well-structured for agents.
Verify before you commit
Use this page to sanity-check Slack quickly. We surface the evidence tier, freshness, and failure posture here, then put the official links where you can actually act on them, especially on mobile.
Evidence
Assessed
Docs reviewed · Mar 11, 2026
Freshness
Updated 2026-03-11T15:12:00+00:00
Mar 11, 2026
Failures
2
2 active failure modes
| Dimension | Score | Bar |
|---|---|---|
| Execution Score Measures reliability, idempotency, error ergonomics, latency distribution, and schema stability. | 7.8 | |
| Access Readiness Score Measures how easily an agent can onboard, authenticate, and start using this service autonomously. | 6.4 | |
| Aggregate AN Score Composite score: 70% execution + 30% access readiness. | 7.2 | |
Agent fails on specific operations (e.g., searching messages) because it uses bot token where user token is required. Error message doesn't indicate token type mismatch.
Agent impact: Agent fails on specific operations (e.g., searching messages) because it uses bot token where user token is required. Error message doesn't indicate token type mismatch.
Workaround: Maintain both token types. Check method documentation for required token type. Map operations to token types at initialization.
Agent hits rate limit on chat.postMessage while other methods are fine. Global rate limiting strategies don't work — must be per-method.
Agent impact: Agent hits rate limit on chat.postMessage while other methods are fine. Global rate limiting strategies don't work — must be per-method.
Workaround: Implement per-method rate tracking. Use Slack's Retry-After header. Prioritize critical messages (alerts) over bulk operations (channel history).
Published review summaries with trust provenance attached to each card.
Docs-backed Built from public docs and product materials.
Test-backed Backed by guided testing or evaluator-run checks.
Runtime-verified Verified from authenticated runtime evidence.
Runtime-observed evidence from 31 proxy interactions. 3 fresh observations within the last 24h. Average confidence: 0.95. Evidence type: latency snapshot. Verified via Rhumb proxy traffic against live Slack endpoints.
Runtime-observed evidence from 38 proxy interactions. 4 fresh observations within the last 24h. Average confidence: 0.96. Evidence type: credential lifecycle. Verified via Rhumb proxy traffic against live Slack endpoints.
Fresh depth-10 runtime rerun passed for Slack auth.test through Rhumb Resolve. The safe proxy auth identity read matched direct Slack control on workspace URL/name, workspace id, bot user identity, bot id, and enterprise-install flag.
Fresh depth-9 runtime rerun passed for Slack auth.test through Rhumb Resolve. The safe proxy auth identity read matched direct Slack control on workspace URL/name, workspace id, bot user identity, bot id, and enterprise-install flag.
Fresh depth-8 runtime rerun passed for Slack auth.test through Rhumb Resolve. The safe proxy auth identity read matched direct Slack control on workspace URL/name, workspace id, bot user identity, bot id, and enterprise-install flag.
Fresh runtime pass verified Slack proxy parity on 8 response fields from POST /api/auth.test: ok, url, team, team_id, user, user_id, bot_id, is_enterprise_install. All matched exactly between Rhumb-managed and direct Slack control.
A fresh internal runtime-review agent with only Slack service access hit Rhumb's Slack proxy on POST /api/auth.test and received the same workspace, user, and bot identity that direct Slack control returned with the same credential. That confirms the current Slack proxy, auth injection, grant path, and runtime evidence pipeline are healthy for a safe non-mutating check.
For Humans 1. Go to https://api.slack.com/apps → Create New App 2. Choose From scratch, name it (e.g., "Rhumb Agent"), select workspace 3.
Value ------- ~200ms ~500ms ~1.5s 99.9%+ Varies by method 90-day history, 10 integrations --- - Idempotency: Messages are NOT idempotent — posting twice creates two messages. Agents must track message timestamps (ts) to avoid duplicates.
Slack is the dominant team communication platform, providing channels, direct messages, threads, and integrations. For agents, Slack serves primarily as an output channel — posting notifications, alerts, reports, and interactive messages. The platform also supports input via slash commands, events API (reacting to messages), and workflow... Reviewed from official documentation.
REST API (Web API) - Base URL: https://slack.com/api - Auth: Bearer token (Authorization: Bearer xoxb-... for bot tokens, xoxp-...
Slack is the dominant team communication platform, providing channels, direct messages, threads, and integrations. For agents, Slack serves primarily as an output channel — posting notifications, alerts, reports, and interactive messages.
Trust shortcuts
This score is documentation-derived. Treat it as a docs-based evaluation of API design, auth, error handling, and documentation quality.
Read how the score works, how disputes are handled, and how Rhumb scored itself before launch.
Overall tier
7.2 / 10.0
No alternatives captured yet.