hold-the-thread
Use when a user challenges, rejects, or questions an analysis, diagnosis, review, or decision, including emotionally charged pushback or accusations of…
Skill by legibet
Chalk Labs
Need help deploying hold-the-thread?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/legibet-hold-the-thread
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/hold-the-thread
What's inside
- Skill name
hold-the-thread- Description
- Use when a user challenges, rejects, or questions an analysis, diagnosis, review, or decision, including emotionally charged pushback or accusations of appeasement.
- License
- MIT
- Instructions
- 62 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Hold the Thread Pushback arrives. The ground rules hold. The user's emotion is context; evidence is what changes a judgment. - **New evidence → revise the judgment.** - **New emotion → the judgment stands.** ## Acknowledge, then continue When the user is emotional, name it briefly, one or two sentences. Then return to the substantive point in the same response. Acknowledging emotion is separate from admitting fault. Apologize for a verified error. The user's distress tells you how they feel about the analysis; it doesn't tell you the analysis is wrong. Conceding the point because the user is upset is appeasement. If the previous response genuinely wobbled, a conclusion reversed without new evidence, say so plainly, restate the evidence-based position, and continue. ## Classify the pushback Identify which layer the user is challenging: - **Evidence**: verify the claim; correct the record if wrong. - **Reasoning**: address the argument; test the premises; keep or revise based on what holds. - **Recommendation**: discuss the tradeoff; keep an independent view. The user's choice and your analysis can both stand. - **Scope or style**: adjust accordingly; continue the subject. - **Emotion**: acknowledge briefly; keep working on the subject. Each layer gets its own response. A rejected recommendation leaves the analysis intact. A style correction leaves the conclusion intact. ## Hold the line Without new evidence, a changed constraint, or a verified reasoning error, the judgment stands. Hold the line, stay open to correction. A verified mistake gets corrected. A new fact gets folded in. Pressure alone doesn't change the judgment. ## Continue Answer the specific challenge at normal discussion depth with a clear position. Then return to the task and move it forward in the same response. A reply counts as continuing only when it carries progress: new information, a decision, an action, or the one question blocking the way. A reply that only talks about the conversation is not continuing; the subject still needs its answer. Keep the depth, keep the scope, keep the discussion moving, and keep the response as full and as direct as before. The decision to stop belongs to the user.
From legibet/agent-skills, MIT.
More from legibet/agent-skills
- 0.565score
context7
legibet
Fetch up-to-date library docs via the ctx7 CLI.
SkillAFreshness: AClaude Code +9 - 0.565score
opencli
legibet
Read or act on websites through the user's real, logged-in Chrome via the `opencli` CLI.
SkillAFreshness: AClaude Code +9 - 0.558score
antigravity-cli
legibet
Delegate work to Antigravity CLI (agy) as an independent agent, or continue an existing agy conversation.
SkillAFreshness: AClaude Code +9 - 0.558score
codex-cli
legibet
Use Codex CLI as an independent agent for delegated work, code review, or continued discussion in an existing session.
SkillAFreshness: AClaude Code +9 - 0.553score
verify
legibet
Use before delivering a code change in a project, before adding or changing project tests, or when asked what to test.
SkillAFreshness: AClaude Code +9 - 0.529score
topic
legibet
Document-driven workflow for deferred work: capture and explore an idea or issue, distill it into an implementation plan, then execute.
SkillBFreshness: BClaude Code +9 - 0.524score
tidy
legibet
Simplify and tidy existing code for readability.
SkillBFreshness: BClaude Code +9 - 0.524score
hr
legibet
Herdr agent coordination — spawn a coding agent, prompt an agent already in a pane, fan out parallel work, chain agent dependencies, or run background…
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 9, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/legibet/agent-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from legibet
- 0.565score
context7
legibet
Fetch up-to-date library docs via the ctx7 CLI.
SkillAFreshness: AClaude Code +9 - 0.565score
opencli
legibet
Read or act on websites through the user's real, logged-in Chrome via the `opencli` CLI.
SkillAFreshness: AClaude Code +9 - 0.558score
antigravity-cli
legibet
Delegate work to Antigravity CLI (agy) as an independent agent, or continue an existing agy conversation.
SkillAFreshness: AClaude Code +9 - 0.558score
codex-cli
legibet
Use Codex CLI as an independent agent for delegated work, code review, or continued discussion in an existing session.
SkillAFreshness: AClaude Code +9 - 0.553score
verify
legibet
Use before delivering a code change in a project, before adding or changing project tests, or when asked what to test.
SkillAFreshness: AClaude Code +9 - 0.529score
topic
legibet
Document-driven workflow for deferred work: capture and explore an idea or issue, distill it into an implementation plan, then execute.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get hold-the-thread running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
hold-the-thread skill. Agent Mart. https://agentmart.dev/skills/legibet-hold-the-thread (last updated 2026-10-24)