CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

deepline-gtm

GTM leads, enrichment, outreach, Plays, developer signals/Reo.dev. Providers: adyntel,affinity,ai_ark,akta,allegrow,amplemarket,apify,attention,attio,aviato,bettercontact,bigquery,bluesky,bounceban,brave,browserbase,builtwith,buyercaddy,cbinsights,clay,clickhouse,close,cloudflare,contactout,contextdev,crunchbase,crustdata-v3,customer_db,databricks,dataforseo,datagma,dealroom,deepline_ip_to_company,deepline_native,deeplineagent,deepseek,discolike,dnb,edges,emailbison,emailguard,enformion,enigma,exa,findymail,firecrawl,fireflies,firmable,forager,fullenrich,fundable,generic_http,getleads,gong,google_ads_audiences,google_workspace,govfiles,grain,hackernews,harvestapi,heyreach,hubspot,hunter,icypeas,instantly,intercom,ipqs,kaspr,kernel,kickbox,leadmagic,lemlist,limadata,linkedin_ads_audiences,luma,lusha,meerapi,meta_audiences,millionverifier,nooks,notion,openmart,opensosdata,openwebninja,outreach,parallel,particle,peopledatalabs,pitchbook,podscan,polymarket,postgres,predictleads,prospeo,quickenrich,rapidapi,….

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Failed to scan

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./skills/deepline-gtm/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

73%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.

The body is an unusually strong operational skill document: a clear sequenced lifecycle with explicit validation checkpoints, recovery loops, cost gating, and a strict output contract, all written leanly. Its one serious defect is progressive disclosure — all ten referenced markdown files are missing from the bundle while the eight provided reference files and twelve scripts are never mentioned, leaving both navigation and the bulk of the bundle content unreachable. Fixing the link structure would raise this substantially.

Suggestions

Ship the missing linked files (writing-outreach.md and the nine recipes/*.md) or rewrite the 'Specialized recipes' and Monitors sections to point at files that exist.

Link the existing bundle from SKILL.md — e.g. the monitors section should reference references/monitor-contract-reference.md and monitor-operations.md, and Clay conversion should reference scripts/clay-to-deepline.mjs and references/clay-action-mappings.md — so the provided material is discoverable.

Trim the opening platform pitch, which restates the description, and split the dense staleAfterSeconds reuse paragraph in section 6 into a short rule list.

DimensionReasoningScore

Conciseness

The body is dense, imperative, and free of concept explanations Claude already knows — commands, rules, and templates only (e.g. "Save files you link in the user's working directory, never a temp directory"). Minor trimmable spots keep it below anchor 5: the opening pitch line ("a GTM tool routing and automation platform for any ad-hoc or automated GTM task, with thousands of tools...") duplicates the description, and the `staleAfterSeconds` reuse paragraph in section 6 packs many rules into one long block. Well above anchor 3 — there is no real padding.

4 / 5

Actionability

Concrete, runnable guidance throughout: exact command lines ("deepline run <name> --input '{...}' --dry-run", "deepline plays check <file>", "deepline runs export <run-id> --dataset <name> --out <file.csv>"), a complete executable Play example, and a copy-paste approval message template. Below anchor 5 because several key steps rely on placeholders and deferred detail ("<input you have -> result you need>", "filter, join and aggregate them with read-only SQL" via "<suggested SQL>", "follow the printed instructions") rather than concrete examples of the common cases.

4 / 5

Workflow Clarity

A clear numbered lifecycle (1 choose what to run → 2 run it → 3 get results → 4 scope and spend → 5 deliver → 6 automate → 7 feedback) with explicit validation checkpoints and error-recovery loops for batch and paid operations: `--dry-run` and `deepline describe` before first calls, "`deepline plays check <file>` # required before running local source", pilot sampling ("run a small representative sample... Stop for a decision on wrong matches, low coverage or a changed cost"), delivery verification ("verify row count, identity and requested columns... read the file back with a strict reader"), and recovery paths ("After a timeout, reconnect with `deepline runs watch <run-id>`; do not launch the run again"). This matches the anchor-5 pattern of explicit validation with feedback loops.

5 / 5

Progressive Disclosure

Every markdown link in the body is dangling: "[outreach copy](writing-outreach.md)" and the nine "recipes/*.md" links (account-orgchart, build-tam, clay-to-deepline, developer-signals, deepline-monitors, find-qualified-titles, portfolio-prospecting, small-business-prospecting) point to files that do not exist anywhere in the bundle. Meanwhile the provided bundle content — 8 files in references/ (monitor-contract-reference.md, clay-action-mappings.md, run-events.md, etc.) and 12 in scripts/ — is never linked from SKILL.md at all, so it is undiscoverable. Scored against the actual bundle per the rubric guideline, this is anchor 2: navigation is effectively broken and real bundle files are orphaned; not 3 because the problem is not unclear signaling but non-resolving links, and the body itself does have clear section structure keeping it above anchor 1.

2 / 5

Total

15

/

20

Passed

Description

57%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.

The description works primarily as a massive provider-keyword trigger list with a terse task-area prefix, which gives good keyword reach but says almost nothing about what the skill actually does or when to use it. It lacks any explicit 'Use when' guidance and includes generic product names (postgres, notion, bigquery) that create real mis-trigger risk. It is functional as a discovery hint but below the quality of the rubric's good examples.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to find or enrich leads, verify emails, run outreach, or work with any of these providers."

State the 'what' plainly before the keyword list: the skill routes GTM tasks to thousands of tools via the deepline CLI (search, run, monitor, automate).

Drop or qualify generic product names (postgres, bigquery, notion, google_workspace, cloudflare, deepseek) that fire on routine non-GTM requests, or mark them clearly as provider integrations.

DimensionReasoningScore

Specificity

"GTM leads, enrichment, outreach, Plays, developer signals/Reo.dev" names the domain and a few concrete task areas, but the bulk of the description is a several-hundred-entry provider name list ("clay,attio,hubspot,gong,...") rather than actions, and it never states what the skill does with these (route? run? find?). It matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive) — above 2 because real task areas are named, below 4 because a provider catalog is not a list of specific capabilities.

3 / 5

Completeness

There is no "Use when..." clause or equivalent explicit trigger guidance — the provider list is a keyword hint, not a when-statement (the body itself calls it "a bounded discovery hint"), so completeness is capped at 3 per the rubric guideline. The 'what' is present but fragmentary ("GTM leads, enrichment, outreach, Plays, developer signals") and the 'when' is only weakly implied by the provider names, which is exactly anchor 3; not 4 because the when-guidance is nowhere explicit.

3 / 5

Trigger Term Quality

The provider list supplies an unusually broad keyword surface ("clay", "hubspot", "gong", "crunchbase", "hunter", "kickbox", "pitchbook") plus natural task nouns "GTM leads", "enrichment", "outreach" — users naming these would naturally trigger the skill. It falls short of anchor 5 because common task phrasings users actually say ("find email addresses", "prospecting", "verify emails", "build a list") are absent or only implicit via provider names, and above 3 because coverage is far beyond 'some relevant keywords'.

4 / 5

Distinctiveness Conflict Risk

The GTM framing gives it a niche, but the provider list includes generic product names that appear constantly in non-GTM work — "postgres", "bigquery", "databricks", "notion", "google_workspace", "cloudflare", "brave", "deepseek", "apify", "firecrawl" — so it would wrongly trigger for ordinary database, search, or productivity requests. That overlap with many other skills matches anchor 3; not 4 because of the broad-name collision risk, not 2 because the GTM/leads context still dominates.

3 / 5

Total

13

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 9 possibly missing (recipes/account-orgchart.md, recipes/build-tam.md, recipes/clay-to-deepline.md, recipes/deepline-monitors.md, recipes/developer-signals.md, +4 more)

Warning

Total

15

/

16

Passed

Repository

getaero-io/gtm-eng-skills

Path

skills/deepline-gtm/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.