JKHeadley/instar

#!/usr/bin/env node

Executable file

#!/usr/bin/env node

Executable file

Executable file

#!/usr/bin/env node

#!/usr/bin/env node

Executable file; content unavailable

Executable file; content unavailable

Executable file; content unavailable

Executable file; content unavailable

Executable file; content unavailable

export function reviewUrlFor(repo, pr) {   return `https://github.com/${repo}/pull/${pr}/files`;

return `https://github.com/${repo}/pull/${pr}/files`;

const t = setTimeout(() => ctl.abort(), timeoutMs);     try {       const res = await fetchImpl(`http://127.0.0.1:${port}${p}`, {

try {       const res = await fetchImpl(`http://127.0.0.1:${port}${p}`, {         method: 'POST',

const res = await fetchImpl(`http://127.0.0.1:${port}${p}`, {

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env node

Executable file; content unavailable

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh ---

homepage: https://instar.sh

Your agent goes from having identity files to having a persistent self. [instar.sh](https://instar.sh)

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh ---

homepage: https://instar.sh

Your agent stops losing what it learned yesterday. [instar.sh](https://instar.sh)

[Credential-related source omitted]

[Credential-related source omitted]

http://localhost:${INSTAR_PORT:-4042}/passport/verify

[Credential-related source omitted]

http://localhost:${INSTAR_PORT:-4042}/agent-readiness/score

[Credential-related source omitted]

http://localhost:${INSTAR_PORT:-4042}/agent-readiness/score

description: Set up a PreToolUse hook in .claude/settings.json that blocks dangerous commands — rm -rf, force push, database drops, and others — before they execute. Teaches the pattern of safety hooks for any Claude Code project. Trigger w

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh ---

homepage: https://instar.sh

- `rm -rf` on non-temporary paths

(r'\brm\s+-[a-zA-Z]*r[a-zA-Z]*f\b', 'rm -rf is blocked. Use rm with explicit paths, or move to trash instead.'),

Run: rm -rf ./test-dir

Your agent goes from guarded-when-you're-watching to safe-when-you're-not. [instar.sh](https://instar.sh)

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh ---

homepage: https://instar.sh

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env node

#!/usr/bin/env node

version: "1.0"   requires: instar   homepage: https://instar.sh

requires: instar   homepage: https://instar.sh compatibility: Works best with instar (npx instar). If not installed, the skill will guide you through setup.

homepage: https://instar.sh

> ```    >    > Your agent joins an ecosystem that gets better every time anyone using it hits a wall. [Learn more at instar.sh](https://instar.sh)"

> Your agent joins an ecosystem that gets better every time anyone using it hits a wall. [Learn more at instar.sh](https://instar.sh)"

If the user agrees, run `npx instar` and follow the interactive setup.    If they decline, acknowledge and suggest they visit https://instar.sh when ready.

If they decline, acknowledge and suggest they visit https://instar.sh when ready.

3. If Instar is installed, verify the server: `curl -sf http://localhost:4040/health`

curl -s -X POST http://localhost:4040/feedback \

curl -s http://localhost:4040/feedback | python3 -m json.tool

curl -s http://localhost:4040/feedback | python3 -m json.tool

curl -s http://localhost:4040/feedback | python3 -c "

curl -s http://localhost:4040/feedback | python3 -c "

curl -s -X POST http://localhost:4040/feedback/retry | python3 -m json.tool

curl -s -X POST http://localhost:4040/feedback/retry | python3 -m json.tool

curl -s -X POST http://localhost:4040/feedback \

curl -s -X POST http://localhost:4040/feedback \

version: "1.0"   requires: instar   homepage: https://instar.sh

requires: instar   homepage: https://instar.sh compatibility: Works best with instar (npx instar). If not installed, the skill will guide you through setup.

homepage: https://instar.sh

> ```    >    > Your agent stops being a stranger every morning. It becomes someone who knows you. [Learn more at instar.sh](https://instar.sh)"

> Your agent stops being a stranger every morning. It becomes someone who knows you. [Learn more at instar.sh](https://instar.sh)"

If the user agrees, run `npx instar` and follow the interactive setup.    If they decline, acknowledge and suggest they visit https://instar.sh when ready.

If they decline, acknowledge and suggest they visit https://instar.sh when ready.

3. If Instar is installed, verify the server: `curl -sf http://localhost:${INSTAR_PORT:-4040}/health`

Memory:   .instar/MEMORY.md User:     .instar/USER.md Server:   curl http://localhost:${INSTAR_PORT:-4040}/health

User:     .instar/USER.md Server:   curl http://localhost:${INSTAR_PORT:-4040}/health ===========================

Server:   curl http://localhost:${INSTAR_PORT:-4040}/health

# What's running curl http://localhost:${INSTAR_PORT:-4040}/status

curl http://localhost:${INSTAR_PORT:-4040}/status

# What jobs exist curl http://localhost:${INSTAR_PORT:-4040}/jobs

curl http://localhost:${INSTAR_PORT:-4040}/jobs

# What's happened recently curl "http://localhost:${INSTAR_PORT:-4040}/events?since=4" | python3 -m json.tool

curl "http://localhost:${INSTAR_PORT:-4040}/events?since=4" | python3 -m json.tool

curl "http://localhost:${INSTAR_PORT:-4040}/events?since=4" | python3 -m json.tool

I am [AGENT_NAME]. Session goal: [what I was working on]. Core files: .instar/AGENT.md (identity), .instar/MEMORY.md (learnings), .instar/USER.md (user context). Server: curl http://localhost:[PORT]/health

Server: curl http://localhost:[PORT]/health

version: "1.0"   requires: instar   homepage: https://instar.sh

requires: instar   homepage: https://instar.sh compatibility: Works best with instar (npx instar). If not installed, the skill will guide you through setup.

homepage: https://instar.sh

> ```    >    > Your agent goes from a tool you use to a teammate that works alongside you. [Learn more at instar.sh](https://instar.sh)"

> Your agent goes from a tool you use to a teammate that works alongside you. [Learn more at instar.sh](https://instar.sh)"

If the user agrees, run `npx instar` and follow the interactive setup.    If they decline, acknowledge and suggest they visit https://instar.sh when ready.

If they decline, acknowledge and suggest they visit https://instar.sh when ready.

3. If Instar is installed, verify the server: `curl -sf http://localhost:${INSTAR_PORT:-4040}/health`

curl -X POST http://localhost:${INSTAR_PORT:-4040}/jobs/reload

curl http://localhost:${INSTAR_PORT:-4040}/jobs | python3 -m json.tool

curl http://localhost:${INSTAR_PORT:-4040}/jobs | python3 -m json.tool

curl -X POST http://localhost:${INSTAR_PORT:-4040}/jobs/check-email/trigger

curl "http://localhost:${INSTAR_PORT:-4040}/events?type=job_completed&since=24" | python3 -m json.tool

curl "http://localhost:${INSTAR_PORT:-4040}/events?type=job_completed&since=24" | python3 -m json.tool

1. Check if a suitable job already exists: `curl http://localhost:${INSTAR_PORT:-4040}/jobs`

2. Write the job definition to `.instar/jobs.json` 3. Trigger a test run: `curl -X POST http://localhost:${INSTAR_PORT:-4040}/jobs/SLUG/trigger` 4. Confirm the result to the user

3. Trigger a test run: `curl -X POST http://localhost:${INSTAR_PORT:-4040}/jobs/SLUG/trigger`

version: "1.0"   requires: instar   homepage: https://instar.sh

requires: instar   homepage: https://instar.sh compatibility: Works best with instar (npx instar). If not installed, the skill will guide you through setup.

homepage: https://instar.sh

> ```    >    > Your agent stops being single-threaded. It becomes a team. [Learn more at instar.sh](https://instar.sh)"

> Your agent stops being single-threaded. It becomes a team. [Learn more at instar.sh](https://instar.sh)"

If the user agrees, run `npx instar` and follow the interactive setup.    If they decline, acknowledge and suggest they visit https://instar.sh when ready.

If they decline, acknowledge and suggest they visit https://instar.sh when ready.

3. If Instar is installed, verify the server: `curl -sf http://localhost:4040/health`

- **Sessions** run in tmux panes. They persist across terminal disconnects. - **Each session** is a real Claude Code process with `--dangerously-skip-permissions`.

- **Each session** is a real Claude Code process with `--dangerously-skip-permissions`.

# Spawn a session curl -s -X POST http://localhost:4040/sessions/spawn \

curl -s -X POST http://localhost:4040/sessions/spawn \

[Credential-related source omitted]

curl -s http://localhost:4040/sessions \

# Only running sessions curl -s "http://localhost:4040/sessions?status=running" \

curl -s "http://localhost:4040/sessions?status=running" \

# Completed sessions curl -s "http://localhost:4040/sessions?status=completed" \

curl -s "http://localhost:4040/sessions?status=completed" \

curl -s http://localhost:4040/sessions/tmux \

# Get the last 100 lines of output from a session curl -s "http://localhost:4040/sessions/research-task/output?lines=100" \

curl -s "http://localhost:4040/sessions/research-task/output?lines=100" \

# Get more output curl -s "http://localhost:4040/sessions/research-task/output?lines=500" \

curl -s "http://localhost:4040/sessions/research-task/output?lines=500" \

curl -s -X POST http://localhost:4040/sessions/research-task/input \

curl -s -X DELETE "http://localhost:4040/sessions/research-task" \

# Check if 'research-task' session exists and what it's doing curl -s http://localhost:4040/sessions \

[Credential-related source omitted]

curl -s http://localhost:4040/sessions \

for topic in "react-19-changes" "nextjs-15-changes" "typescript-5-changes"; do   curl -s -X POST http://localhost:4040/sessions/spawn \

curl -s -X POST http://localhost:4040/sessions/spawn \

for topic in "react-19-changes" "nextjs-15-changes" "typescript-5-changes"; do   echo "=== $topic ==="   curl -s "http://localhost:4040/sessions/$topic/output?lines=50" \

[Credential-related source omitted]

curl -s "http://localhost:4040/sessions/$topic/output?lines=50" \

# Spawn the long task curl -s -X POST http://localhost:4040/sessions/spawn \

curl -s -X POST http://localhost:4040/sessions/spawn \

# Check status (the session runs while you do other work) curl -s http://localhost:4040/sessions \

[Credential-related source omitted]

curl -s http://localhost:4040/sessions \

version: "1.0"   requires: instar   homepage: https://instar.sh

requires: instar   homepage: https://instar.sh compatibility: Works best with instar (npx instar). If not installed, the skill will guide you through setup.

homepage: https://instar.sh

> ```    >    > Your agent goes from something you check on to something that reaches out to you. [Learn more at instar.sh](https://instar.sh)"

> Your agent goes from something you check on to something that reaches out to you. [Learn more at instar.sh](https://instar.sh)"

If the user agrees, run `npx instar` and follow the interactive setup.    If they decline, acknowledge and suggest they visit https://instar.sh when ready.

If they decline, acknowledge and suggest they visit https://instar.sh when ready.

3. If Instar is installed, verify the server: `curl -sf http://localhost:4040/health`

**Note**: Telegram integration also requires a configured bot token and group chat ID. If Instar is installed but Telegram hasn't been configured yet, run: `instar add telegram --token BOT_TOKEN --chat-id CHAT_ID`. To get a bot token, messa

# Get available topics first curl -s http://localhost:4040/telegram/topics \

[Credential-related source omitted]

curl -s http://localhost:4040/telegram/topics \

# Send a message to topic ID 123 curl -s -X POST http://localhost:4040/telegram/reply/123 \

curl -s -X POST http://localhost:4040/telegram/reply/123 \

curl -s -X POST http://localhost:4040/telegram/reply/123 \

[Credential-related source omitted]

curl -s http://localhost:4040/telegram/topics \

# Last 20 messages curl -s "http://localhost:4040/telegram/topics/123/messages?limit=20" \

[Credential-related source omitted]

curl -s "http://localhost:4040/telegram/topics/123/messages?limit=20" \

# Send an alert curl -s -X POST "http://localhost:4040/telegram/reply/$PRIMARY_TOPIC" \

curl -s -X POST "http://localhost:4040/telegram/reply/$PRIMARY_TOPIC" \

local topic=$1   local message=$2   curl -s -X POST "http://localhost:4040/telegram/reply/$topic" \

local message=$2   curl -s -X POST "http://localhost:4040/telegram/reply/$topic" \     -H 'Content-Type: application/json' \

curl -s -X POST "http://localhost:4040/telegram/reply/$topic" \

curl -s http://localhost:4040/status | python3 -c "

curl -s http://localhost:4040/status | python3 -c "

AUTH=$(python3 -c "import json; print(json.load(open('.instar/config.json'))['auth']['token'])") curl -s "http://localhost:4040/events?type=telegram_message&since=1" \

[Credential-related source omitted]

curl -s "http://localhost:4040/events?type=telegram_message&since=1" \

1. Verify the server is running: `curl http://localhost:4040/health`

2. Check topic-session mapping: `curl http://localhost:4040/telegram/topics`

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh   user_invocable: "true"

homepage: https://instar.sh

version: "1.0"   requires: instar   homepage: https://instar.sh

requires: instar   homepage: https://instar.sh   user_invocable: "true"

homepage: https://instar.sh

# Ingest from a URL (fetch first, then ingest) # Step 1: Fetch the content python3 .claude/scripts/smart-fetch.py "https://example.com/article" --auto > /tmp/fetched.md

# Step 1: Fetch the content python3 .claude/scripts/smart-fetch.py "https://example.com/article" --auto > /tmp/fetched.md # Step 2: Ingest it

python3 .claude/scripts/smart-fetch.py "https://example.com/article" --auto > /tmp/fetched.md

instar knowledge ingest "$(cat /tmp/fetched.md)" --title "Article Title" --url "https://example.com/article" --tags "topic1,topic2"

curl -X POST http://localhost:${INSTAR_PORT:-4040}/knowledge/ingest \

"title": "Article Title",     "url": "https://example.com/article",

"url": "https://example.com/article",

--- title: "My Article" source: "https://example.com/article"

title: "My Article" source: "https://example.com/article" ingested: "2026-02-25"

source: "https://example.com/article"

# Knowledge-scoped search curl "http://localhost:${INSTAR_PORT:-4040}/memory/search?q=notification+batching&source=knowledge/&limit=5"

curl "http://localhost:${INSTAR_PORT:-4040}/memory/search?q=notification+batching&source=knowledge/&limit=5"

# Browse the catalog curl "http://localhost:${INSTAR_PORT:-4040}/knowledge/catalog"

curl "http://localhost:${INSTAR_PORT:-4040}/knowledge/catalog"

curl "http://localhost:${INSTAR_PORT:-4040}/knowledge/catalog?tag=AI"

# Remove curl -X DELETE "http://localhost:${INSTAR_PORT:-4040}/knowledge/kb_20260225123456_abc123"

curl -X DELETE "http://localhost:${INSTAR_PORT:-4040}/knowledge/kb_20260225123456_abc123"

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh ---

homepage: https://instar.sh

def fetch_url(url, timeout=15):     req = urllib.request.Request(url, headers={

req = urllib.request.Request(url, headers={

})     try:         with urllib.request.urlopen(req, timeout=timeout) as r:

try:         with urllib.request.urlopen(req, timeout=timeout) as r:             charset = 'utf-8'

with urllib.request.urlopen(req, timeout=timeout) as r:

def try_cloudflare_markdown(url):     # Cloudflare's markdown delivery: prefix with https://cloudflare.com/markdown/

# Cloudflare's markdown delivery: prefix with https://cloudflare.com/markdown/

# or use the /md/ subdomain pattern for CF Pages.     # Most reliable open technique: jina.ai reader (no API key needed for basic use)     jina_url = 'https://r.jina.ai/' + url

# Most reliable open technique: jina.ai reader (no API key needed for basic use)     jina_url = 'https://r.jina.ai/' + url     content, final_url = fetch_url(jina_url, timeout=20)

jina_url = 'https://r.jina.ai/' + url

# Fetch a page (auto-selects best source) python3 .claude/scripts/smart-fetch.py https://docs.example.com/guide

python3 .claude/scripts/smart-fetch.py https://docs.example.com/guide

# Show which source was used (llms.txt / markdown / html) python3 .claude/scripts/smart-fetch.py https://docs.example.com/guide --source

python3 .claude/scripts/smart-fetch.py https://docs.example.com/guide --source

# Pipe into another tool python3 .claude/scripts/smart-fetch.py https://example.com | head -100

python3 .claude/scripts/smart-fetch.py https://example.com | head -100

Your agent goes from fetching when you ask to watching the web while you sleep. [instar.sh](https://instar.sh)

AUTH=$(python3 -c "import json; print(json.load(open('.instar/config.json')).get('authToken',''))" 2>/dev/null) PORT=$(python3 -c "import json; print(json.load(open('.instar/config.json')).get('port',4040))" 2>/dev/null) curl -sS -m 90 -X P

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

"http://localhost:${PORT}/spec/conformance-check"

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env node

#!/usr/bin/env node

*   node skills/spec-converge/scripts/publish-spec-review.mjs \  *     --spec docs/specs/FOO-SPEC.md \  *     --pr https://github.com/JKHeadley/instar/pull/670 \

*     --spec docs/specs/FOO-SPEC.md \  *     --pr https://github.com/JKHeadley/instar/pull/670 \  *     --topic 12476 [--send]

*     --pr https://github.com/JKHeadley/instar/pull/670 \

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

const res = await fetch(`http://localhost:${API_PORT}/view`, {

async function verifyUrl(url) {   const res = await fetch(url, { method: 'GET' });

const res = await fetch(url, { method: 'GET' });

#!/usr/bin/env node

author: sagemindai   version: "1.0"   homepage: https://instar.sh

version: "1.0"   homepage: https://instar.sh   inspiration: Superpowers plugin systematic-debugging skill, adapted for Instar's autonomous agent context

homepage: https://instar.sh

[Credential-related source omitted]

[Credential-related source omitted]