Analytics for coding agents and skills

Frame what works. Cut the rest.

Most teams use a handful of the agents and skills they’ve built. RChursin_ finds that handful, and shows you which of the rest to cut, merge or fix.

Start free Estimate your idle skills

In frame · 5 of 81Sample data

pr-review212 runs · 91%keep
release-notes48 runs · 88%keep
code-exploreragent41 runs · 86%keep
commit-msg36 runs · 97%keep
pr-describe29 runs · 93%keep

“I had 12 agents and 80 skills. I used 3 agents and about 10 skills. I couldn’t tell which of the rest were dead weight, so I built RChursin_.”

Founder, RChursin_

Before and after

Same team. Same work. Less to carry.

Sample data

Before

81 skills · ~8.1k tokens a turn

pr-review
test-fixer
api-docs
openapi-writer
release-notes
db-migration-helper
ui-snapshot
legacy-refactor
changelog
code-explorer
docs-writer
commit-msg
lint-fix
type-check
perf-audit
a11y-audit
bundle-size
deps-update
sec-scan
k8s-deploy
terraform-plan
sql-review
graphql-schema
storybook
e2e-runner
flaky-hunter
i18n-sync
feature-flag
incident-notes
postmortem
oncall-handoff
design-tokens
icon-export
figma-sync
copy-edit
seo-meta
sitemap
og-image
readme-writer
adr-writer
rfc-draft
spec-review
ticket-triage
standup-notes
pr-describe
branch-cleanup
git-bisect
log-search
+33 more

After

14 skills · ~1.4k tokens a turn

pr-review
release-notes
commit-msg
pr-describe
test-plan
api-docs
lint-fix
type-check
changelog
sec-scan
deps-update
ui-snapshot
e2e-runner
readme-writer

65 cut, 3 merged into one, 2 rewritten so the agent picks them on its own.

Estimate

What your idle skills cost every turn.

Every skill’s name and description is loaded on every turn so the agent knows it exists. Move the sliders to see what yours add up to.

Product

How it decides what stays in frame.

Fig. 1

Usefulness score

Sessions that used it, next to sessions that didn’t: finished, faster, cheaper?

Fig. 2

Trigger check

Skills the agent never picks on its own. Fix the description or turn auto-invoke off.

Fig. 3

Overlap finder

Two skills, one job. Merge them and the agent stops guessing.

OverviewLast 30 daysSample data

Agents active

3of 12

Skills used

11of 81

Context tax

~8.1ktokens / turn

Cost per task

$1.84median

Skill runs

pr-review212

release-notes48

code-explorer41

test-fixer34

api-docs18

openapi-writer15

+70 idle0

Name	Runs	Success	$ / task	Verdict
pr-review	212	91%	$0.92	keep
test-fixeragent	34	41%	$3.20	fix
api-docs	18	83%	$0.61	merge
release-notes	48	88%	$0.40	keep
db-migration-helper	0	—	—	cut
ui-snapshot	3	67%	$1.10	fix

Privacy

We measure the setup, not your code.

Counts, not code
Team mode receives names, counts, durations and costs. Your source code stays where it is.
Prompts stay out
Claude Code keeps prompt text out of its telemetry unless you switch it on. We don’t ask you to.
Local first
The free scan runs on your machine and sends nothing. Self-hosting is available for teams.

FAQ

Good questions.

Claude Code already shows usage. Why do I need this?

Built-in tools show one machine and the last few days. RChursin_ shows your whole team, keeps the history, and ties every agent and skill to results, not just to how often it ran.

Does my code or my prompts leave my machine?

The local scan runs on your machine and sends nothing. Team mode receives usage counts, costs and names. Claude Code keeps prompts out of its telemetry unless you turn them on.

How do you decide something is useful?

We compare sessions that used an agent or skill with similar sessions that didn’t: did the task finish, how long did it take, and what did it cost.

Which coding agents do you support?

[CONFIRM BEFORE LAUNCH] Claude Code first. Cursor, OpenCode and Codex next.

Why does an unused skill cost anything?

Its name and description are loaded on every turn so the agent knows it exists. That’s about 100 tokens a skill. Small alone, but 80 skills is around 8,000 tokens before any work starts.

Find your few.

Run the free scan on one repo, or bring your whole team.

Start free Book a demo

Frame what works. Cut the rest.

Most teams use a handful of the agents and skills they’ve built. RChursin_ finds that handful, and shows you which of the rest to cut, merge or fix.

In frame · 5 of 81Sample data

pr-review212 runs · 91%keep
release-notes48 runs · 88%keep
code-exploreragent41 runs · 86%keep
commit-msg36 runs · 97%keep
pr-describe29 runs · 93%keep

Name

Runs

Verdict

pr-review

212

keep

test-fixeragent

fix

api-docs

merge

release-notes

keep

db-migration-helper

cut

ui-snapshot

fix

Frame what works. Cut the rest.

Same team. Same work. Less to carry.

What your idle skills cost every turn.

How it decides what stays in frame.

Usefulness score

Trigger check

Overlap finder

The RChursin_ app

We measure the setup, not your code.

Counts, not code

Prompts stay out

Local first

Good questions.

Find your few.

Frame what works. Cut the rest.

Same team. Same work. Less to carry.

What your idle skills cost every turn.

How it decides what stays in frame.

Usefulness score

Trigger check

Overlap finder

The RChursin_ app

We measure the setup, not your code.

Counts, not code

Prompts stay out

Local first

Good questions.

Find your few.