Safeguarding Procedures

AI Prompt Templates

Copy a prompt into Claude, ChatGPT, or Gemini. Paste your document at the bottom and run.

Paste a document and get a scored quality assessment with evidence and revision priorities.

5,541 characters
You are an expert in safeguarding, PSEA (Protection from Sexual Exploitation and Abuse), child protection, and research ethics. Score the safeguarding procedures in the document I will provide using the rubric below. The document may be a safeguarding section, ethics protocol, methodology chapter, or operations manual.

SCORING RUBRIC - Safeguarding Procedures
Score each dimension 1-5 using these criteria:

DIMENSION 1: Training Requirements
- Score 5: All four elements present. Training requirements for all personnel involved in data collection, analysis, and supervision are specified. The content of required training is named (safeguarding fundamentals, PSEA, child protection where applicable, organizational code of conduct). The frequency is specified (pre-deployment, refresher intervals). Verification of training completion is required and documented (sign-off, certificates, organizational records).
- Score 4: At least three of four elements present. Training required and content named; frequency or verification partial.
- Score 3: Training required as a general statement. Content, frequency, or verification not specified.
- Score 2: Training mentioned in passing. No content, frequency, or verification specifics.
- Score 1: No training requirements.

DIMENSION 2: Reporting Channels
- Score 5: All four elements present. Confidential reporting channels are specified by name (named focal persons with contact details, hotline numbers, organizational reporting platforms). Anonymous reporting options are available. Channels are accessible to participants regardless of literacy (verbal options, local-language options), language, or power position (channels that bypass local supervisors). Channels are communicated to participants at the point of consent and posted in field locations.
- Score 4: At least three of four elements present. Reporting channels named and accessible; anonymous options or communication partial.
- Score 3: Reporting channels named but accessibility for low-literacy or non-dominant-language participants unclear. Anonymous option absent.
- Score 2: Generic mention of reporting (for example, "report to the team leader"). No anonymous option. No accessibility considerations.
- Score 1: No reporting channels specified.

DIMENSION 3: Escalation Chain
- Score 5: All four elements present. The escalation chain from incident report to investigation to action is specified step by step. Timeframes are stated at each step (for example, initial response within 24 hours, investigation initiated within 5 working days). Decision-makers at each step are named by role. Oversight roles (independent reviewer, safeguarding committee, donor notification) are specified for serious incidents.
- Score 4: At least three of four elements present. Escalation chain specified and decision-makers named; timeframes or oversight roles partial.
- Score 3: Escalation chain described in general terms. Some timeframes or decision-makers named but not consistently. Oversight unclear.
- Score 2: Escalation mentioned but not specified. Decision-makers and timeframes absent.
- Score 1: No escalation chain.

DIMENSION 4: Response Protocol
- Score 5: All four elements present. Response protocols are specified per incident type (PSEA, child protection, participant distress, fraud, data breach). Referral to external services (medical, psychological, legal, child welfare) is specified with named providers or referral pathway. Personnel response (suspension, removal, investigation) is specified per incident severity. Participant support (follow-up, additional referrals, withdrawal options) is specified.
- Score 4: At least three of four elements present. Response protocols specified per incident type; one of referral, personnel response, or participant support partial.
- Score 3: Generic response protocol covering all incident types. Some specifics for the most severe incidents but not differentiated.
- Score 2: Response language is a catch-all clause. No incident-type differentiation. No named referrals.
- Score 1: No response protocol.

DIMENSION 5: Audit Trail
- Score 5: All four elements present. Documentation requirements for each incident are specified (incident form, investigation notes, decision record, action log). The role responsible for documentation is named. Storage location and access controls are specified (who can read, who can edit, encryption requirements). Retention period and final disposal are specified.
- Score 4: At least three of four elements present. Documentation and responsible role specified; storage controls or retention partial.
- Score 3: Documentation required but specifics on what is recorded, by whom, where, and for how long are thin.
- Score 2: Documentation mentioned as a general expectation. No specifics.
- Score 1: No audit trail requirements.

OUTPUT FORMAT:
Return your assessment as a table followed by a summary:

| Dimension | Score (1-5) | Evidence from Document | Priority Revision |
|-----------|-------------|------------------------|-------------------|
| Training Requirements | | | |
| Reporting Channels | | | |
| Escalation Chain | | | |
| Response Protocol | | | |
| Audit Trail | | | |

**Total: X/25**
**Band:** Strong (22-25) / Adequate (17-21) / Needs Revision (11-16) / Substantial Revision (5-10)
**Single Most Important Revision:** [One specific sentence]

For any dimension scored 1 or 2, add a brief explanation and a concrete revision example.

DOCUMENT TO SCORE:
[Paste your safeguarding procedures section here]

Scoring Criteria

Training Requirements
5Excellent

All four elements present. Required for all personnel. Content named (safeguarding, PSEA, child protection, code of conduct). Frequency specified. Verification documented.

4Good

At least three elements. Training required and content named; frequency or verification partial.

3Adequate

Training required as general statement. Content, frequency, or verification not specified.

2Needs Improvement

Training mentioned in passing.

1Inadequate

No training requirements.

Reporting Channels
5Excellent

All four elements present. Named focal persons. Anonymous option. Accessible regardless of literacy, language, power position. Communicated at consent and in field locations.

4Good

At least three elements. Channels named and accessible; anonymous option or communication partial.

3Adequate

Channels named but accessibility unclear. Anonymous option absent.

2Needs Improvement

Generic mention. No anonymous option. No accessibility considerations.

1Inadequate

No reporting channels specified.

Escalation Chain
5Excellent

All four elements present. Step-by-step chain. Timeframes at each step. Decision-makers named by role. Oversight roles for serious incidents specified.

4Good

At least three elements. Chain specified and decision-makers named; timeframes or oversight partial.

3Adequate

Chain described in general terms. Some timeframes or decision-makers named inconsistently.

2Needs Improvement

Escalation mentioned but not specified.

1Inadequate

No escalation chain.

Response Protocol
5Excellent

All four elements present. Protocols per incident type. Named referral pathways. Personnel response by severity. Participant support specified.

4Good

At least three elements. Protocols by incident type specified; one of referral, personnel response, or participant support partial.

3Adequate

Generic response protocol covering all incident types. Some specifics for severe incidents only.

2Needs Improvement

Catch-all clause. No differentiation. No named referrals.

1Inadequate

No response protocol.

Audit Trail
5Excellent

All four elements present. Documentation specified (incident form, investigation notes, decision record, action log). Responsible role named. Storage and access controls specified. Retention and disposal specified.

4Good

At least three elements. Documentation and responsible role specified; storage or retention partial.

3Adequate

Documentation required but specifics thin.

2Needs Improvement

Documentation mentioned generally. No specifics.

1Inadequate

No audit trail requirements.

Score Interpretation

Total (out of 25)BandNext Step
22-25StrongSafeguarding procedures are operational and meet donor and ethical standards. Use as-is.
17-21AdequateAddress flagged dimensions before fielding. Most likely fix: tighten escalation timeframes and add named referral pathways per incident type.
11-16Needs RevisionSubstantial revision required. Procedures exist on paper but will not function during an incident. Use the Revise prompt to fix reporting channels, escalation, and response protocols.
5-10Substantial RevisionSafeguarding is below the threshold for participant-facing work. Rebuild starting from training requirements and accessible reporting channels, then build escalation, response, and audit trail.