Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/crisng95/flowkit/fk-refresh-urlsgit clone --depth 1 https://github.com/crisng95/flowkitWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.01129 |
| Opus 5 | $0.00000 | $0.00564 |
| Sonnet 5 | $0.00000 | $0.00226 |
| Haiku 4.5 | $0.00000 | $0.00113 |
Grade C, and why
fk-refresh-urls scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
PID=$(curl -s "http://127.0.0.1:8100/api/videos/${VID}" | python3 -c "import sys,json; print(json.load(sys.stdin)['project_id'])") Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s http://127.0.0.1:8100/api/flow/status How it starts
The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Refresh expired GCS signed URLs for all scenes in a video (images, videos, upscale videos) and character reference images.
Usage: /fk-refresh-urls <video_id> [--project-id <PID>]
When to use
- Before
/fk-review-videoif videos were generated hours ago (GCS signed URLs expire) - Before
/fk-concat-fit-narratorif downloading from URLs instead of local files - After any long gap between generation and consumption of media URLs
Pre-flight
# Extension must be connected with flow key
curl -s http://127.0.0.1:8100/api/flow/status
# Must show: {"connected": true, "flow_key_present": true}
# If flow_key_present is false: open/refresh a Google Flow tab in Chrome
Step 1: Get project_id from video
VID="<video_id>"
PID=$(curl -s "http://127.0.0.1:8100/api/videos/${VID}" | python3 -c "import sys,json; print(json.load(sys.stdin)['project_id'])")
echo "Project: $PID"
Step 2: Bulk refresh via TRPC
This calls Google Flow's TRPC flow.getFlow endpoint, extracts ALL fresh signed URLs from the response, and updates scenes + characters in DB.
curl -s -X POST "http://127.0.0.1:8100/api/flow/refresh-urls/${PID}" | python3 -c "
import sys, json
r = json.load(sys.stdin)
print(f\"Refreshed: {r.get('refreshed', 0)} URLs (found {r.get('found', 0)} total)\")
if r.get('error'):
print(f\"ERROR: {r['error']}\")
"
What gets updated:
horizontal_image_url/vertical_image_url— scene imageshorizontal_video_url/vertical_video_url— scene videos (original)horizontal_upscale_url/vertical_upscale_url— 4K upscaled videosreference_image_url— character/entity reference images
The server matches each URL's media_id against *_media_id fields on scenes and characters, updating whichever orientation/type matches.
Step 3: Verify refresh worked
# Check a few scenes have valid URLs
curl -s "http://127.0.0.1:8100/api/scenes?video_id=${VID}" | python3 -c "
import sys, json
scenes = sorted(json.load(sys.stdin), key=lambda s: s['display_order'])
# Auto-detect orientation
ori = 'horizontal'
for s in scenes:
if s.get('horizontal_video_status') == 'COMPLETED' and s.get('horizontal_video_url'):
ori = 'horizontal'; break
if s.get('vertical_video_status') == 'COMPLETED' and s.get('vertical_video_url'):
ori = 'vertical'; break
ok = 0; expired = 0
for s in scenes:
url = s.get(f'{ori}_video_url') or ''
if url and 'Expires=' in url:
import re, time
m = re.search(r'Expires=(\d+)', url)
if m and int(m.group(1)) > time.time():
ok += 1
else:
expired += 1
elif url:
ok += 1
else:
expired += 1
print(f'Orientation: {ori.upper()}')
print(f'Valid URLs: {ok}/{len(scenes)}')
if expired:
print(f'Still expired: {expired} — may need to open Flow tab in Chrome for flow key')
else:
print('All URLs refreshed successfully!')
"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 117 lines · 0 tokens per session scan C ee3358394373
fk-refresh-urls is a command published in the GitHub repository crisng95/flowkit (620 stars, last pushed 14d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,129 tokens. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.