catalan-adobe/slack-cdp
Control Slack via CDP or headless API tokens. Navigate channels, read/send messages, search conversations, check unreads, and manage status. Two modes: CDP (Slack desktop with --remote-debugging-port) for full UI control, or headless (xoxp/xoxb token) for data operations without Slack running. Triggers on: slack, read slack, search slack, slack unreads, send slack message, slack status, navigate slack, check slack, slack messages, go to channel, slack DM.
72
91%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Critical
Do not install without reviewing
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that covers all evaluation dimensions well. It provides specific concrete actions, comprehensive trigger terms, clear 'what' and 'when' guidance, and is distinctly scoped to Slack operations. The description of two operational modes adds useful context for skill selection without being overly verbose.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: navigate channels, read/send messages, search conversations, check unreads, manage status. Also describes two distinct operational modes (CDP and headless) with clear details about each. | 3 / 3 |
Completeness | Clearly answers both 'what' (control Slack via CDP or headless API tokens with specific actions listed) and 'when' (explicit 'Triggers on:' clause with comprehensive trigger terms). Both dimensions are thoroughly covered. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms explicitly listed: 'slack', 'read slack', 'search slack', 'slack unreads', 'send slack message', 'slack status', 'navigate slack', 'check slack', 'slack messages', 'go to channel', 'slack DM'. These are terms users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche around Slack specifically. The two-mode architecture (CDP vs headless) and Slack-specific trigger terms make it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, highly actionable skill with excellent workflow clarity—the decision tree, chaining patterns, verification steps, and error recovery table are all well-executed. The main weakness is that it's somewhat long for a single file, with the token setup and headless API examples adding bulk that could be offloaded to supporting files. Conciseness could improve by trimming the OAuth token acquisition steps and trusting Claude to know basic curl patterns.
Suggestions
Move the 'Obtaining tokens' section (OAuth flow details) to a separate TOKENS.md reference file to reduce SKILL.md length
Trim the headless API curl examples—Claude knows how to call REST APIs; a table of endpoint + required params would be more token-efficient than full curl commands
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient but includes some sections that could be tightened—the token obtaining instructions are quite detailed for something Claude could infer, and the headless API curl examples are somewhat verbose. However, most content earns its place given the dual-mode complexity. | 2 / 3 |
Actionability | Fully executable bash commands throughout, concrete curl examples for headless mode, specific data-qa selectors for clicking, and copy-paste ready chaining patterns. The script discovery block and all command invocations are directly executable. | 3 / 3 |
Workflow Clarity | Clear decision tree for choosing the right tool, explicit verification step (always use 'where' after 'navigate'), well-sequenced chaining patterns with multi-step examples, error recovery table, and a clear headless-vs-CDP decision framework. The navigate→where feedback loop is explicitly called out. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and headers, but it's a long monolithic file (~180 lines) where the headless token setup instructions and detailed curl examples could be split into a separate reference file. No bundle files are provided to offload content to. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
Reviewed
Table of Contents