pipeline-health-reviewer — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited pipeline-health-reviewer (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
This skill diagnoses why a B2B pipeline is stalling and delivers a prioritized set of structural fixes. It treats pipeline problems as systems problems, not sales performance problems, unless the evidence points clearly to the latter.
This skill works best with specific data. Ask the user to provide as much of the following as possible:
If no data is provided, run the diagnostic as a structured interview, asking for the key numbers before delivering conclusions.
A stalling pipeline has one of four root causes. Identify which one or which combination applies before recommending fixes.
Deals appear to move but do not actually progress because stages are defined by time or activity rather than buyer behavior.
Signals:
The fix: Redefine stages around buyer actions, not sales actions. A deal should not move to "Proposal" because the rep sent a proposal. It should move when the buyer has confirmed they received it, reviewed it, and have a specific next step scheduled. This is the distinction between deal stage theater and real pipeline visibility.
Pipeline is full but close rates are low because deals that should not be in the pipeline are in the pipeline. This inflates the number, destroys forecast accuracy, and wastes rep time on deals that will never close.
Signals:
The fix: Audit closed-lost deals for patterns. If a certain company size, vertical, tech stack, or buyer persona consistently loses, that is an ICP problem. The qualification framework needs to include a hard disqualification step, not just a scoring model that everyone ignores. Better to have 30 real deals than 90 deals where 60 are noise.
Deals are real and qualified but stalling because there is no structured follow-up after key milestones. The buyer goes quiet and the rep waits.
Signals:
The fix: Every deal that advances past a defined stage needs a follow-up sequence with a specific trigger and a specific message, not a generic nudge. The sequence should reference the last thing the buyer said, the specific problem they described, and the next concrete action. Deals with no next step scheduled should be flagged automatically and escalated.
The sales cycle is long because the buying process requires multiple stakeholders, budget approval, or security review that takes time regardless of how well the deal is being worked. This is a real constraint, not a pipeline problem.
Signals:
to competitors
The fix: This is a pipeline design problem, not a sales execution problem. The answer is to build the long cycle into the forecast model and stop treating it as a stall. Separately, identify which stages in the process can be compressed: can security review start earlier, can legal review a standard agreement in parallel with negotiation, can the champion be given materials to pre-sell internally.
Output in this format:
PIPELINE HEALTH REVIEW
[Company or team name if provided]
Review date: [today's date]
PIPELINE SNAPSHOT
[Restate the key numbers the user provided. If numbers are missing,
note what is needed to complete the diagnosis.]
PRIMARY DIAGNOSIS
[One of the four root causes above, or a combination. Be direct.
Do not hedge if the signals point clearly to one cause.]
EVIDENCE
[The specific data points or descriptions that support the diagnosis.
No speculation. Only what the user actually described.]
VELOCITY BREAKDOWN BY STAGE
[If stage data was provided: which stages have normal velocity,
which are stalling, and what the likely cause is for each stall.]
RECOMMENDED FIXES (in priority order)
1. [First fix -- the highest leverage change]
What to change: [specific]
Expected impact: [what this should do to velocity or close rate]
Time to implement: [realistic estimate]
2. [Second fix]
What to change: [specific]
Expected impact: [specific]
Time to implement: [estimate]
3. [Third fix]
What to change: [specific]
Expected impact: [specific]
Time to implement: [estimate]
WHAT NOT TO DO
[The common wrong answers for this type of pipeline problem. If
the diagnosis is stage definitions, do not hire more reps. If it
is ICP, do not run more campaigns. Name the wrong moves explicitly.]
30-DAY LEADING INDICATOR
[One metric to watch in the next 30 days that will tell the user
whether the fixes are working. Not a lagging metric like closed
revenue. A leading indicator like average days in stalling stage
or number of deals with next steps scheduled.]data or assume details about their sales motion.
what is needed.
change how the pipeline is designed, not to add more activity.
A pipeline full of bad deals needs to be said plainly.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.