CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

chat-connectors

Connect a ChatGPT or Claude account and sync its conversation history into the brain automatically. Cookie paste-in is the primary lane; sync is incremental (watermark + trailing-window gap-heal), and can run on a schedule via autopilot or host cron. Distinct from the export-file lane (that lives in conversation-archive) — this is the LIVE, account-connected, auto-scraping path.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Medium

Suggest reviewing before use

Fix and improve this skill with Tessl

tessl review fix ./plugin/skills/chat-connectors/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced operational guide with executable commands, explicit validation checkpoints, honest failure handling, and sharp dedup boundaries against sibling skills. The one real weakness is redundancy: the partial/forbidden/watermark semantics are restated in five different sections, inflating token cost without adding information.

Suggestions

State the 'partial → watermark does not advance → re-run' and 'forbidden → pivot to export-file lane' semantics once (e.g. in Troubleshooting) and reference them elsewhere instead of repeating them in First Sync, Contract, When it fails, and Anti-Patterns.

Consider moving the Contract section's guarantee list into a reference file (or merging it with When it fails / Anti-Patterns) to cut the body's token footprint while keeping the guarantees discoverable.

DimensionReasoningScore

Conciseness

The body is dense with skill-specific detail rather than generic explanation, but the same guidance is repeated across sections — 'partial → watermark did not advance, re-run' appears in First Sync, Troubleshooting, Contract, When it fails, and Anti-Patterns, and the 'forbidden → export-file lane' pivot appears four times.

3 / 5

Actionability

Copy-paste-ready commands cover every common case: 'gbrain connectors auth chatgpt --cookie -', 'gbrain connectors sync chatgpt --dry-run / --limit 5 / --full', 'gbrain conversation-parser scan conversations/chatgpt/<slug>', a host cron line, and an env-var escape hatch (GBRAIN_CONNECTOR_<PROVIDER>_COOKIE).

5 / 5

Workflow Clarity

The batch-import workflow is clearly sequenced with explicit validation checkpoints and feedback loops: auth ends in a probe verdict, then dry-run → --limit 5 sample → 'conversation-parser scan' validation of one page → full backfill, with defined recovery for 'partial' (re-run, watermark holds), 'forbidden' (pivot lanes, never loop-retry), and health surfacing in 'gbrain doctor'.

5 / 5

Progressive Disclosure

No bundle files exist, and all cross-references (conventions/, sibling SKILL.md files) are one level deep and clearly signaled with a conventions block up top; sections are well organized, though ~208 lines with the Contract and repeated troubleshooting content could be trimmed or split.

4 / 5

Total

17

/

20

Passed

Description

70%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A specific, well-bounded description with concrete capabilities and an explicit distinction from the export-file sibling skill. Its main gaps are the missing 'Use when...' trigger guidance inside the description itself and some internal jargon ('watermark + trailing-window gap-heal', 'brain') that a user would not naturally use.

Suggestions

Add an explicit 'Use when...' clause to the description field itself (e.g. 'Use when the user asks to connect their ChatGPT/Claude account or keep their chat history synced') rather than relying on the separate when_to_use frontmatter field.

Replace or gloss internal jargon such as 'watermark + trailing-window gap-heal' and 'the brain' with plain terms (e.g. 'only fetches new conversations and heals gaps') so capabilities read as user-facing actions.

Include one or two natural synonyms users would say, such as 'auto-import my chats' or 'download my conversations', to broaden trigger coverage.

DimensionReasoningScore

Specificity

Lists several concrete actions ('Connect a ChatGPT or Claude account and sync its conversation history', 'Cookie paste-in is the primary lane', 'run on a schedule via autopilot or host cron'), but internal jargon like 'watermark + trailing-window gap-heal' and 'the brain' keeps it from being comprehensively plain-language.

4 / 5

Completeness

The 'what' is explicit and concrete, but the description text itself contains no 'Use when...' clause or equivalent trigger guidance (the when_to_use/triggers live in separate frontmatter fields), which caps completeness at 3.

3 / 5

Trigger Term Quality

Natural terms users would say are present ('ChatGPT', 'Claude account', 'sync', 'conversation history'), but a few common variations are missing (e.g. 'import my chats', 'download my conversations').

4 / 5

Distinctiveness Conflict Risk

Explicitly draws the boundary against the sibling skill: 'Distinct from the export-file lane (that lives in conversation-archive) — this is the LIVE, account-connected, auto-scraping path', giving it a clear niche with minimal conflict risk.

5 / 5

Total

16

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 9 suspicious

Warning

Total

14

/

16

Passed

Repository

garrytan/gbrain

Path

plugin/skills/chat-connectors/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.